University of Tennessee, Knoxville TRACE: Tennessee Research and Creative Exchange
Masters Theses Graduate School
12-1998
Handling Qualities Evaluation of a Variable Stability Navion Airplane (N66UT) using Frequency Domain Test Techniques
Randy Lee Bolding University of Tennessee - Knoxville
Follow this and additional works at: https://trace.tennessee.edu/utk_gradthes
Part of the Aerospace Engineering Commons
Recommended Citation Bolding, Randy Lee, "Handling Qualities Evaluation of a Variable Stability Navion Airplane (N66UT) using Frequency Domain Test Techniques. " Master's Thesis, University of Tennessee, 1998. https://trace.tennessee.edu/utk_gradthes/1572
This Thesis is brought to you for free and open access by the Graduate School at TRACE: Tennessee Research and Creative Exchange. It has been accepted for inclusion in Masters Theses by an authorized administrator of TRACE: Tennessee Research and Creative Exchange. For more information, please contact [email protected]. To the Graduate Council:
I am submitting herewith a thesis written by Randy Lee Bolding entitled "Handling Qualities Evaluation of a Variable Stability Navion Airplane (N66UT) using Frequency Domain Test Techniques." I have examined the final electronic copy of this thesis for form and content and recommend that it be accepted in partial fulfillment of the equirr ements for the degree of Master of Science, with a major in Aviation Systems.
William D. Lewis, Major Professor
We have read this thesis and recommend its acceptance:
Ralph Kimberlin, Fred Stellar
Accepted for the Council: Carolyn R. Hodges
Vice Provost and Dean of the Graduate School
(Original signatures are on file with official studentecor r ds.) To the Graduate Council:
I am submitting herewith a thesis written by RandyLee Bolding entitled "Handling Qualities Evaluation of a Variable Stability Navion Airplane(N66UT) using Frequency Domain Test Techniques." I have examinedthe final copy of this thesis for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Master of Science, with a major in Aviation Systems.
WilliamD. Lewis,
We have read this thesis
Accepted for the Council:
Associa��-te Vice Chancellorand � Dean of the Graduate School HANDLING QUALITIES EVALUATION OF A VARIABLE STABILITY NAVION AIRPLANE (N66UT) USING FREQUENCY -DOMAIN TEST TECHNIQUES
A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville
Randy Lee Bolding December 1998 DEDICATION
This thesis is dedicated to my wife
Michelle, whose love, support and encouragement
give the strength for all of my accomplishments.
To my children, Joshua and Savannah, for the sacrificeof
a father's time.
ii ACKNOWLEGEMENTS
I would like to express my gratitude to Dr. William Lewis, Dr. Ralph Kimberlin, and Mr. Fred Stellar fo r their assistance in the preparation of this document. In addition,
I would like to thank Charles Catterall fo r endless hours of computer programming and data reduction, and thanks to the airport staff in the preparation and instrumentation of the test aircraft.
111 ABSTRACT
General aviation small aircraft handling quality certification has predominately been accomplished using traditional time-domain test techniques. This thesis investigates the handling quality characteristics of the variable stability Navion airplane, tail number
N66UT, using frequency-domain test techniques. N66UT is configured with conventional flight controls at the copilot's station and a fly-by-wire set of flight controls at the pilot's station. Time delays between the fly-by-wire and conventional flight controls were determined to be minimal. Evaluation of handling qualities of the aircraft were compared to the fixed wing flying qualities specified in MIL-HDBK- 1797.
Analysis of flight test was conducted using Comprehensive Identification from
Frequency Responses (CIFER) program developed at NASA Ames Research Center,
Moffe tt Field, California.
Frequency-domain system-identification methods are well suited to aircraft flight control and handling-qualities analysis since many current design specifications, design and analysis techniques, and acceptance flight-test techniques are based in the frequency domain. The response characteristics of the Navion N66UT were stable fo r the longitudinal, lateral and rudder control inputs. The handling quality parameters met or exceeded proposed Level 1, Category C requirements, and Level 2, Category A requirements fo r fixed-wing military aircraft. The fly-by-wire system closely mirrored conventional control behavior well beyond bandwidth frequency. Time delay differences between conventional and FBW systems ranged from very near 0 to 9.5 ms. Throttle response data results were inconclusive. Further flight-testing is required.
IV TABLE OF CONTENTS
CHAPTER PAGE
I. INTRODUCTION 1
II. BACKGROUND 3
Flight Dynamics And Handling Qualities 3
Literature Review (Specifications) 5
Mll..-F-8785C 5
Mll...-HDBK- 1797 6
III. FLIGHT TEST TECHNIQUE 13
Description of Test Aircraft 13
General Description 13
Conventional Flight Controls 16
Fly-By-Wire Flight Controls 19
Conventional Engine Control System 28
Fly-By-Wire Engine Control System 32
Instrumentation, Calibrations And Corrections 32
Test Method And Procedures 34
Special Precautions 38
Data Collection 38
IV. DATA REDUCTION TECHNIQUE 42
v TABLE OF CONTENTS (continued)
v. DISCUSSION OF TEST RESULTS 53
Data Analysis 53
General 54
Longitudinal Control Input 55
Pitch Mode, Conventional Control System 55
Pitch Mode, Conventional Control System Doublet 57
Pitch Mode, FBW Control System 57
Pitch Mode, Conventional And FBW
Control Systems 58
Pitch Mode To Normal Acceleration 58
Pitch Mode, Cross-Coupling 59
Lateral Control Input To Roll Attitude Change 59
Roll Mode, Conventional Controls 59
Roll Mode, Conventional Controls Doublet 60
Roll Mode, FBW Control System 60
Roll Mode, Conventional And FBW
Control System 61
Adverse Yaw 61
Roll Mode Lateral Acceleration (AY) 62
vi TABLE OF CONTENTS (continued)
Pedal Control Input To Yaw Attitude Change 63
Yaw Mode, Conventional Controls 63
Yaw Mode, FBW Control System 65
Yaw Mode, Conventional And FBW
Control System 66
Dihedral Effect 66
Yaw Mode, Lateral Acceleration (AY) 67
Throttle Control Input Test Results 67
VI. CONCLUSIONS 69
VII. RECOMMENDATIONS 74
REFERENCES 76
BIBLIOGRAPHY 79
APPENDICES 83
Appendix A: Navion Calibration Charts 84
Appendix B: Bode Plots 104
VITA 123
VII LIST OF TABLES
TABLE PAGE
3-1 Navion Specifications 14
3-2 Servo Response 22
3-3 Inputs Of Moment Controls 27
3-4 Inputs To Normal Force Control 29
3-5 Inputs To Thrust/Drag Modulation System 29
3-6 Tests And Test Conditions 35
6-1 Summary Of Tests Results 71
6-2 Time Delays in the Fly-By-Wire Control System with
Respect to Conventional Controls 71
viii LIST OF FIGURES
FIGURE PAGE
2-1 Cooper-Harper Handling Qualities Rating Scale 7
2-2 Bandwidth Requirements 10
3-1 Ryan Navion N66UT 13
3-2 Navion Three-View 15
3-3 Conventional System Control Column 17
3-4 Conventional Rudder Control System 18
3-5 Conventional Aileron Control System 20
3-6 Conventional Elevator Control System 21
3-7 Fly-By-Wire Rudder Block Diagram 23
3-8 Fly-By-Wire Rudder And Elevator Diagram 24
3-9 Fly-By-Wire Aileron control System 26
3-10 Engine Throttle Control System 30
3-11 N66UT Engine Controls And Instrumentation 31
3-12 Fly-By-Wire Computer Interface 33
3-13 Flight Test Weight And Balance Information 36
3-14 IoTech Data Acquisition System 39
ix LIST OF FIGURES (continued)
4-1 Frequency Response Method For System Identification 43
4-2 Time History: Longitudinal Stick Input To Pitch Rate Response 46
4-3 Conventional Controls Longitudinal Input To Pitch Attitude
Response Auto Spectrum 47
4-4 Typical CIFERBo de Plot. A Comparison Of The Fly-By-Wire
And Conventional Control Systems 52
4-5 Definitions Of Bandwidth And Phase Delay 51
5-1 Longitudinal Time History (Conventional Controls);
Longitudinal Input To Pitch Response 56
5-2 Conventional Control Pedal Input To Lateral Acceleration
Time History 64
A-1 Copilot's Airspeed Instrument Error Correction 85
A-2 Airspeed Position Error Correction 86
A-3 Copilot's Altimeter Instrument Error Correction 87
A-4 FBW Aileron Stick Position (CHOO) 88
A-5 FBW Elevator Stick Position (CHOl) 89
A-6 FBW Rudder Pedal Position (CH02) 90
A-7 FBW Throttle Command Position (CH03) 91
A-8 Roll Rate (p) (CH04) 92
A-9 Pitch Rate (q) (CHOS) 93
X LIST OF FIGURES (continued)
A-1 0 Yaw Rate (r) (CH06) 94
A-ll Longitudinal Acceleration (ax) (CH07) 95
A-12 Lateral Acceleration (ay) (CH08) 96
A-13 Normal Acceleration (az) (CH09) 97
A-14 Aileron Surface Position (CH 1 0) 98
A- 15 Elevator Surface Position (CHll) 99
A-16 Rudder Surface Position (CH12) 100
A-17 Std. Throttle Control Position (CH13) 101
A-18 Roll Attitude (<(>) (CH14) 102
A- 19 Pitch Attitude (8) (CHIS) 103
B-I Conventional Control Longitudinal Input To Pitch
Attitude Response 105
B-2 Conventional Control Longitudinal Doublet 106
B-3 FBW Control Longitudinal Input To Pitch
Attitude Response 107
B-4 Conventional And FBW Control Longitudinal Input To
Pitch Attitude Response 108
B-5 Longitudinal Control Input To Heave 109
B-6 Longitudinal Control Input To Roll Response IIO
B-7 Conventional Control Lateral Input To Roll Response Ill
xi LIST OF FIGURES (continued)
B-8 Conventional Control Lateral Doublet 112
B-9 FBW Control Lateral Input To Roll Response 113
B-10 Lateral Control Input To Roll Response 114
B-11 Lateral Control Input To Yaw Response 115
B-12 Lateral Control Input To Lateral Acceleration 116
B-13 Conventional Control Pedal Input To Yaw Response 117
B-14 FBW Control Pedal Input To Yaw Response 118
B-15 Pedal Control Input To Yaw Response 119
B-16 Pedal Control Input To Roll Response 120
B-17 Pedal Control Input To Lateral Acceleration 121
B-18 Throttle Control Input To Longitudinal Acceleration 122
xii NOMENCLATURE
� Difference (delta)
Or Change in Rudder Deflection
<1> Roll Attitude
Loa Rolling Moment due to Change in Aileron Deflection
M& Pitch Moment due to Change in Elevator Deflection ms Milliseconds
Nor Yaw Moment due to Change in Rudder Deflection
8 Pitch Attitude
COsw Bandwidth Frequency co135 Frequency at 45 degree Phase Margin
C01so Frequency at 180 degree Phase
'tp Phase Delay
x-axis Acceleration due to Change in Throttle X01
Yaw Attitude \jl
Abbreviations
AID Analog to Digital
Accel Acceleration
AFDD Aeroflightdynamics Directorate
xiii Abbreviations (continued)
ax x-axis acceleration ay y-axis acceleration az z-axis acceleration
C.G. Center of Gravity
C-H Cooper-Harper
CIFER Comprehensive Identification from Frequency Response
COMPOSITE Multi-window Spectral Analysis dB Decibels
Deg Degree
Deg/s2 Degrees per second squared
DFC Direct Force Control
Degrees Celsius
FBW Fly By Wire
FCS Flight Control System
FRESPID Frequency Response Identification ft Feet
FW Fly-by-Wire
FWD Forward
HD Density Altitude
Hz Hertz
xiv •
Abbreviations (continued)
lAS Indicated Air Speed
In. Inch
Kbytes Kilobytes
KCAS Knots Calibrated Air Speed
KHz Kilo-hertz
lbs Pounds
MAC Mean Aerodynamic Chord
MIMO Multi-input Multi-output
MISO Multi-input Single output
MISOSA Multi-input Single-output Spectral Analysis
NASA National Aeronautics and Space Administration
OAT Outside Air Temperature
p Roll Rate
PC Personal Computer
PIO Pilot Induced Oscillation
q Pitch Rate
r Yaw Rate rad/sec Radians/second rpm Revolutions-Per-Minute
sec Second
XV Abbreviations (continued)
SISO Single-input Single-output
ST CIFERDesig nation for Conventional Flight Controls
Std Conventional Flight Controls
U.S. United States
UTSI University of Tennessee Space Institute wt Weight
xvi Chapter I
INTRODUCTION
In July 1991 , the University of Tennessee Space Institute (UTSI) purchased two
Navion aircraft from Princeton University. These aircraft contain several modifications
for conducting flight research funded by the Department of the Navy. Each aircraft is
equipped with a fly-by-wire (FBW), variable control response-feedback system utilizing
electronically signaled, hydraulically actuated controls located at the copilot station, and
conventional flight controls at the pilot station. The FBW flight control system allows
direct independent control over the aircraft six degrees of freedom through a direct force
control (DFC) flight computer. The fly-by-wire system was designed and assembled
using individual components purchased and integrated to form a variable stability system.
Differences may exist between the fly-by-wire and conventional flight controls due to
control implementation techniques.
These aircraft provide an excellent platform for performance, stability, and
control test instruction for the University's Flight Research Department. For these
aircraft to provide a suitable flight test platform using the variable stability systems, a
thorough documentation of the basic handling characteristics of the aircraft configured
with null settings for all controls is required. Data gathered from basic aircraft handling qualities tests can then be used to further investigate control responses in an attempt to
model several variable settings. The responses due to various settings provide the
1 capability to replicate several different type aircraft flight characteristics using a single platform.
The purpose of this thesis is to evaluate the handling quality characteristics of the
Ryan Navion, tail number N66UT, using frequency-domain test techniques. Short-term responses to control inputs are the primary focus for this thesis. The investigation explored differences between the two control station responses. Evaluation of aircraft handling qualities derived from the Navion are compared to the fixed wing flying qualities outlined in MIL-HDBK-1797 [1].
Analysis of frequency domain data is conducted using the Comprehensive
Identification from frEquency Response (CIFER®) (2] program developed at NASA
Ames Research Center, Moffett Field, California. CIFER has been used extensively for helicopter and engine analysis, but not applied to fixed wing aircraft.
2 Chapter II
BACKGROUND
FLIGHT DYNAMICS AND HANDLING QUALITIES
For the majority of flight dynamics investigations time domain analysis is usually
adequate, particularly when the test aircraft is a classical, unaugmented airplane. When
the test airplane is an advanced modern airplane fitted with an advanced control system,
flight dynamics analysis in the frequency domain can provide additional, valuable insight
into its behavior. For an aircraft configured with a fly-by-wire system, the control system becomes an integral part of the primary signal flow path and the influence of its dynamic
characteristics on fl ying and handling qualities is of critical importance. [3]
The flying and handling qualities of an airplane are those properties that describe the ease and effectiveness with which it responds to commands in the execution of some flight task. When handling qualities are described qualitatively and are formulated in terms of pilot opinion, they tend to be somewhat subjective. Flight dynamics involves the relatively short-term motion of an airplane in response to control input or to an external disturbance such as atmospheric turbulence. The motion of interest can vary from small excursions about trim to very large amplitude inputs for maneuvering when normal aerodynamic behavior may become very non-linear. The dynamic behavior of an airplane is significantly shaped by its stability and control properties, which in turn have their roots in the aerodynamics of the airframe. Modernflight dynamics is concernednot
3 only with the dynamics, stability and control of the basic airframe but also with interaction between airplane and flight controlsystems. [3]
A measure of the handling qualities of an aircraft is its stability margin when operated in a closed-loop-tracking task. The maximum frequency at which a closed-loop tracking can take place without degrading response characteristics is known as bandwidth
( rosw ). Aircraft capable of operating at a sufficiently large value of bandwidth will have superior performance when responding to disturbances. A bandwidth criterion is especially useful for highly augmented aircraft in which the response characteristics have a non-classical form. The concept of using bandwidth for flying qualities assessment has been proven. A 1970 utilization of bandwidth was a part of Neal-Smith [4] criterion consisting of empirical bounds on the closed-loop pitch attitude resonance I 9/9c I max versus pilot equalization for a piloted closure designed to achieve a specifiedbandwidth.
Experience with this criterion has shown that the results can be sensitive to the selected value of closed-loop bandwidth. [5]
System identification is a procedure for accurately characterizing the dynamic response behavior of a complete aircraft, subsystem, or individual component from measured data. Frequency-domain identification approaches are especially well suited to the development and validation of flight-control systems. Feedback stability and noise amplification properties are determined from the open-loop frequency response, and characterized by metrics such as crossover frequency, and associated gain and phase margins. Command tracking performance is determined from the closed-loop frequency response, and characterized by metrics such as bandwidth and time-delay. Frequency-
4 domain identification approaches allow the direct and rapid identification of these frequency responses and metrics, without the need to first identify a parametric (state space) model structure such as is required in applying time-domain methods.
The availability of comprehensive and reliable computational tools has substantially enhanced the acceptability of frequency-domain techniques in the flight control and flight-test communities. Benefits from applying these techniques include the reduction of flight-test time required for control system optimization and handling qualities evaluation, especially for complex control-law architectures, and improvements in the fi nal system performance. Frequency-domain methods offer a transparent understanding of component and end-to-end response characteristics that can be critical in solving system integration problems. [6]
LITERATURE REVIEW (Specifications)
MIL-F-8785C [7].
MIL-F-8785C specifications contain the requirements for the flying and handling qualities, in flight and on the ground, of U.S. Military, manned, piloted airplanes. It is intended to assure flying qualities that provide adequate mission performance and flight safety regardless of design implementation or flight control system mechanization. [5]
This specification defines requirements for the conventional aircraft, as tested using traditional time-domain test techniques. Testing the variable-stability Navion, an aircraft with direct force controls (DFC) which allow independent control over the six inertial degrees of freedom, exceeds the scope of MIL-F-8785C.
5 MIL-HDBK-1797.
MIL-STD-1797 A has been redesignated as a handbook, and is to be used for guidance purposes only. The document is no longer to be cited as a requirement. The only physical change from MIL-STD-1797A was the cover page. This handbook is intended for use with fixed-wing aircraft, including a category pertaining to the Navion.
The MIL Standard and Handbook are presented in terms of aircraft response or control axes. This change from MIL-F-8785C is designed to accommodate highly augmented aircraft, which is a primary objective of MIL-HDBK-1797 [1]. Within MIL-HDBK-
1797, a wide variety of closed-loop tasks have been developed for the evaluation of aircraft flyingqualities. Most tasks evaluated within the handbook are based on historical data and research experiments with ground-based and in-flight simulators. The handling characteristics described within the handbook are specified in terms of qualitative degrees of suitability and levels.
6 Adequacy for SelectedTask orR equired Alman Demandson the Pilot In Selected Pilot Operation* Character15tla Task or RequiredOperation' Rating
Excellent Pilot compensation not a factor for desired I Highly desirable performance
Good Pilot compensationnot a factor for desired .. 2 ... Negligible deficiencies performance 1 YES Fair Some mildly Minimal pilot compensation required for 3 unpleasant deficiencies desired performance
Minor but annoying Desiredperformance requires moderate 4 !sit deficiencies pilot compensation Deficiencies satisfactory Moderatelyobjectionable Adequate performance requires considerable warrant NO 5 without deficiencie1 pilot compensation .. improvement improvement .. Veryobjectionable but Adequateperformance requires extensive 6 tolerabledeficiencies pilot compensation
J�YES Adequateperformance not auainable witb Major deficiencies muimum tolerablepilot compensation. 7 Is adequate Controllabilitynot in question Deficiencies performanceattainable NO require with a tolerable pilot Major defiCiencies Considerable pilot compensationis 8 ... improvement workload ... requiredfor control
Majordeficiencies Intensepilot compensation required is to 9 retaincontrol • YES Is it Controlwill lost duringsome portion Improvement Major deficiencies be 10 NO.., controllable? manditory of�ired �ation .. L .J1
• Defmitionof required operationinvol vesdesignation of flight Pilot• Decisions phasesand/or sub phaseswith accompanying conditions
Figure 2-1. Cooper-HarperHandling Qualities Rating Scale [ 6]
7 The degrees of suitability are defined using the Cooper-Harper (C-H) rating scale (Figure
2-1) as follows:
• Satisfactory: Flying qualities clearly adequate for the mission Flight Phase. Desired
performance is achievable with no more than minimal pilot compensation. (C-H
ratings from 1 through 3)
• Acceptable: Flying qualities adequate to accomplish the mission Flight Phase, but
some increase in pilot workload or degradation in mission effectiveness, or both,
exists. (C-H ratings from 4 through 6)
• Controllable: Flying qualities such that the aircraft can be controlled in the context of
the mission Flight Phase, even though pilot workload is excessive or mission
effectiveness is inadequate, or both. The pilot can transition from Category A Flight
Phase tasks to Category B or C Flight Phases, and Category B and C Flight Phase
tasks can be completed. (C-H ratings from 7 through 9)
Level 1 is Satisfactory, Level 2 is Acceptable, and Level 3 is Controllable.
The criteria in the quantitative open-loop requirements are based on interpolation between and extrapolation beyond the configurations, flight conditions, and tasks that have been evaluated in the existing database. Pilot handling qualities ratings have been consolidated and boundaries established based on bandwidth and phase margins. It is normally recommended that more than one evaluation be conducted for each test condition, using between three to six evaluation pilots for time-domain flight testing techniques. With the use of frequency domain test techniques and specifications, one
8 instrumented flight has the ability to capture the required data with very little subjectivity
placed on the pilot.
The bandwidth of the open-loop pitch attitude response to pilot control force is
recommended to be within the bounds shown in Figure 2-2. Bandwidth (OOsw) is the highest frequency at which either the response of aircraft pitch attitude to pilot control force and control-deflection inputs have 45 degrees or more of phase margin (ro135) or 6 dB of gain margin, whichever comes first. Individual specifications for uncoupled roll, yaw, and throttle axis responses are not presented in the handbook. For the purpose of this thesis, Figure 2-2 specifications will be used for all axis evaluations.
9 0.2
0.15
� 0 0 � 0.1 Q) � 0. 1-' .c� � 0.05
0 0 2 4 6 8 10 1 2 Bandwidth
ro8w (rad/sec)
(a) Category A Requirements
0.2
0.1 5
>. <:I ,.-.. 0 (.) Q) 0 "' 0.1 Q) '-' c. 1-' .c� � 0.05
0 0 0.5 1 .5 2 2.5 3 3.5 4 4.5 5 Bandwidth
ro8w (rad/sec)
(b) Category C Requirements
Figure 2-2. Bandwidth Requirements [1].
10 To use Figure 2-2 charts, determine whether the response being evaluated is designated a
Category A or Category C maneuver. Enter the appropriate chart at the bandwidth limit
(x-axis). Move up until reaching the phase delay value associated with this bandwidth limit (y-axis). Read the Level designation assigned to this combination of values.
Flight Phase Definitions:
Non-terminal Flight Phases:
Category A: Those non-terminal Flight Phases that require rapid maneuvering, precision tracking, or precise flight-path control. Included in this category are:
• Air-to-air combat (CO)
• Ground attack (GA)
• Weapon delivery/launch (WD)
• Aerial recovery (AR)
• Reconnaissance (RC)
• In-flight refueling (receiver) (RR)
• Close formation flying (FF)
• Low-altitude parachute extraction (LAPES) delivery
Category B: Those non-terminal Flight Phases that are normally accomplished using gradual maneuvers and without precision tracking, although accurate flight-path control may be required. Included in this category are:
• Climb (CL)
• Cruise (CR)
• Loiter (LO)
• In-flight refueling (tanker) (RT)
• Descent (D)
• Emergency descent (ED)
• Emergency deceleration (DE)
• Ariel delivery (AD)
11 Terminal Flight Phases:
Category C: Terminal Flight Phases are normally accomplished using gradual maneuvers and usually require accurate flight-path control. Included in this category are:
• Takeoff (TO)
• Approach (PA)
• Waveoff/go-around (WO)
• Landing (L)
12 Chapter III
FLIGHT TEST TECHNIQUE
DESCRIPTION OF TEST AIRCRAFf
General Description
The aircraft used for this flight test was a Navion, Serial Number NAV-42013,
Registration Number N66UT, manufactured by Ryan Aeronautical Company, San Diego,
California. (Figure 3-1) The Ryan Navion is a low-wing four-place, dual-control
airplane powered by a single air-cooled engine. The fu selage is an all-metal one-piece
semi-monocoque structure. The tail unit is a cantilever monoplane type with detachable
tips. There is a two piece tailpane and elevators that are interchangeable left and right.
Normal fuel capacity is 40 U.S. Gallons. A general description of the Navion is contained in reference 8. Aircraft specifications are listed in Table 3-1. A Navion three view is presented in Figure 3-2. N66UT is owned and maintained by the University of
Tennessee Space Institute, Tullahoma, Tennessee.
Figure 3-1. Ryan Navion N66UT
13 Table 3-1. Navion Specifications
Aircraft: Length ...... 27 ft, 6 in. (8.38 m.) Wing: Span (b)...... 33 ft, 5 in. (10.18 m.) Area (S)...... 184 ft2 (17.1 12 m2.) Sweep, leading edge ...... 2.996 deg. Aspect Ratio (A) ...... 6.04
Taper Ratio (A.) ...... 0.54 Mean aerodynamic chord ...... 5.7 ft (1.74 m.) Dihedral ...... 7.5 de g. Airfoil: Tip...... NACA 6410 Root ...... NACA 44 15 2 Wing loading ...... 14.6 lb./sq. ft. (71.2 kg./m. ) Horizontal tail: Area ...... 43 ft2 ( 4.0 m?) Aspect ratio ...... 4.0 Taper ratio ...... 0.67 Airfoil ...... NACA 0012
Vertical tail: Area above horizontal stabilizer ...... 15.5 ft2 (1.44 m?) Rudder area ...... 8.33 ft2 (0.78 m.2)
14 Figure 3-2. Navion Three-view
15 N66UT has several modifications made by Princeton University under various
United States Navy Grants that contrast this aircraft from the conventional Navion. The engine is a Teledyne-Continental I0-520B producing 285 take-off horsepower at sea level, 2700 RPM. The propeller is a McCauley three-bladed constant speed propeller, part number D3663/582NC-2, serial number K103943YS. The conventional Navion main landing gear struts are replaced with those designed for the Camair twin (a Navion conversion with nearly 40 percent increase in gross weight). N66UT landing gear is permanently fixed in the down (extended) position. The system can be adjusted for landing tests to provide and allowable sink rate of 12.5 feet per second.
Conventional Flight Controls
The conventional flight control system is located at the copilot's station. This system consists of a set of rudder pedals, a wheel for elevator and aileron control, and cables and linkage connected to the respective control surfaces. The control column, to which the control wheel shafts (through universal joints) are attached, pivots at the base to permit fore and aft movement (Figure 3-3). Sprockets on the forward end of each control wheel shaft are interconnected by a chain, the ends of which attach to cables routed through pulleys at the top and bottom of the column. The control wheel and shaft on the left side have been removed. The rudder pedal assembly is hinged to the floor
(Figure 3-4). The conventional rudder pedal assembly at the leftpilot's station has been removed. The rudder control system consists of two cable assemblies, connected to rudder pedal torque tube arms, and running aft to the rudder horns. Two rods, extending forward from the pedals to the bellcrankfor nose wheel steering, serve as a balanceca ble
16 DET�l A Q . Remove STEP1 bolt, andthen dide controt wheel '\:)ondtubeoutthroughcontrolponel
A �l2Atoft 0 leq) AN3M4Nut 0 Req) DET�l B ) AH2(.11A&It (21eq) · AH36S4Nut (2 Req 1 � 2 Removebolts securing rudderto pedak \Jvertical onm, and then remove pedals.
Figure 3-3. Conventional System Control Column [8]
17 DErAIL A
· 1 Secure pedals together, fnQYO the two ;>edol• full IO 5 Check rudder pedal travel. Make sure pedal ho rujlers do root hit floor befpre rudder oHoifts f.,U travel. ,uu. . B . RUOOER $YSTUol FOIIWAIIO BELI.CIIAH� AND NOS£ GEAR STEERING M£CHAN15M 2 Jock oitplone. line nose wheel with ceftlerline of oirplofte, oftd then odju•t steering bellcroftk rollers so thot.rollers ore 01 do.. to point contoct on &frut orm 01 pouible. without ptelooding rollen ogoinst the orm !Fore Oftd oft positioft of rollers cor. be changed by shimming bellcronk 1upport brodet with woshen ond hoff· woshers.l N 0 TE: To align ,.....,wheel. level oirplonlt,drop plumb bob• � from engine cronk,hoh and wing center ..ction, draw o chalk line between the plumb bol». and set wheel potOIIelto line Adju1l cobl.- to position rud· 4 der 3 d-srees to the right of streornli"" po»ition.ond tight· en to 30pounds tensioft. A. AN2�19 Bolt 12 Req1 AN96().416 Washer (2Req) AN24-13 Bolt(2 Req ) AN321).4 Nul (4 Req) AN381H-3 Coller 14 Reql L AN2J.IO Bolt 12 R�ql ANJI0-3 Nut (2 �� Check now wheel travel with ply. 6 AN960-10L Washer 12 Req) p Figure 3-4. Conventional Rudder Control System [8] 18 for the system. The two rudder-to-aileron coordinating cables and spring assemblies have been removed from N66UT. The ailerons are controlled by a combination linkage and cable system (Figure 3- 5). Disconnect fittings are located within the control cable guard box on the pilot's floor, and turnbuckles are located in the right wheel well. Adjustable rods connect the bell cranks to the ailerons. The elevator control system consists of two cable assemblies, connecting the control column arm with the elevator hom (Figure 3-6). Cables pass through the empenage guided by a series of pulleys and connect to the elevator hom. Fly-By-Wire Flight Controls The pilot's station has been equipped with an analog fly-by-wire control system containing power-actuated control surfaces commanded by electrical signals. The signals come from the various cockpit controllers and motion sensors, and when appropriately processed and summed, provide a net signal to each servo-actuator resulting in an airplane response of a particular character and magnitude. The servos are hydraulic, supplied by an engine-driven hydraulic pump delivering approximately 9 gallons-per minute at 750 pounds-per-square-inch pressure. Control of pitch, roll, and yaw are through conventional elevator, aileron, and rudder control surfaces. The full authority of each surface is available and the maximum deflection rate is approximately 70 degrees per second. The hydraulic servos are modified Bendix units originally designed for the Convair B-58 Hustler and incorporate built-in solenoids and pilot force-override disengage features. Servo response parameters are presented in Table 3-2. 19 NOTE: Rvddor- 2 S.Curo control wh110ls in neutral pooition. Thon with cross ovor ccblos slack. til)!.lon c.-.blos from control column to bollcronks unt il both oil.oron� oro approximately 2 degrees obove neutral. 3 Tighten crosso""r free control wh110ls ond ch..:lc for 25 dog,.... up and 17-112 desJroeo down trovel. Minor corrections o<• me>do by reodjwing turnbuckles; h-.... ro- odju&tment of push-pull rods may bo neces�ry. 5 Sotbellctonk stops to clear bolkronl �· ....� ....,·��------" � Coble Tension 30 pounds • Surface Travel :ZSO up, 17-112"clown l T tnvel T oleronce •2" ��...... ,...... �-----....a With telohod aileron 25 c!o:grooo up, A AN21�A Pulley (2 Roql f AN3-U Bolt (4 Req) adjust push-pull rod• on each side AN� i!.olt (1 Req) AN310.3 Nut (4 Roql 10 that oft cdblo attachment AN31� Nut Req) AN960-10L o er Roq) bollcronk arm clocrs pu•h-pull rod (1 W oh (4 inch. AN:Jro-2-3 Conet (l Req) AN3SI}-2-2 Cottor (4 Roq) by 118 Spocet 8 AN21�A Pulloy (2 Roq) .¢5.3-10.19 {4 Reql AN47 Bolt (2 Reql· 14S..S230l Rollo< (t Req) Check eNite m AN310.4 Nut (2 Reql ayote for 30 pound• tension. AN23-12 Bolt Roq) 6 ANla0-2-3 CoHet (2 Roq) G (4 AN310.3 Nut (4 Roq) C AN3-6 I!Oit (2 Reql AN9�10l Wosher (4 Roq) AN310.3 Nvt (2 Roq) AN380-2-2 Conor (4 Roq) ANJ.90..2-3 Cone< (2 Roq) H AN�9 Bolt C AN210.:iA�OR7-JI• Pulley {1 Req) (2 Ro:J) NOTE: ANI.- Nut I 0 !l.olt Req) AN310.3 {2 ?�) Check c!ign:nontof na, en� t:c::J!;r.•J o...""gc1. (1 AN3ro-C2-:> Cot1or (2 Rcz<;) oiktrc.n AN310.4 N ut (1 Roo) Jf trciang odgosCt'-41 cv;,t 1/4 ir.ch ot.:.:c.fcHz."" • .n".Or.+� AN:l.W-2-2 Conor (l' Roq) I cej::ot ci14ron end i'.cp ·�oing oquclly to align AN�!2 !<:It (4 l:;q) tho 114-inch tolctcnco. E AN210-4A-4>R7-314 Pulloy (21toq) AN31� I-Jut(4 >lo>q) ..dgeo within AN4-10 Bo!l'(2 Req) A.'<�&>-1. !6 Wc,hor (� l!cq) AN310-4 Nut (2 Req) AN�2-2 ;:cxcr (4 Roql AN:l.a0-2-2 Cottet (2 Roq) Figure 3-5. Conventional Aileron Control System [8] 20 1 Sccu:e control colvmns forword so thot ends. of column ore 114 inch from firewoll :.our.dproofing; ond ther. t•ghten cables to 30 povnds tension so thct el�wotoc is ogoins.t the down stop. .• AN3-6 Bolt (2 Fie;) AN31Q.3 Nut (2 Fle.ql AN380-2·2 Colfer (2 Req) B. AN210-4A-4DR7-3:4 Pulley (l Req) AN4-10 8olr (l R"q) AN31Q.4 Nut (1 Req) ANJS0.2-2 Cotter {l Req) AN380-3-4 Cotter (1 Req) C. AN23-12 8olr (2 Reo) AN23-21 Bolt (1 Req) AN320·3 Nut (3 Reg) AN960-10 Wo>her (3 Req) AN3S0-2-2 Corter (3 Req) tl. AN21Q.3A-4DR7-3 '4 Pulley (l Req) AN4.10 Bolt (l Reo) AN310.4 Nut (i Req) AN380-2-2 Cotter (1 Req) 2 Pull control columns oft until elevator hits the up stop; t en pull control wheeh 3 degrees (roughly one inch h AN21Q.3A-tDR7-3.'t Pulley (2 Req) travel) further oft to check control column ouembly E. cle-arance. 'rhis overtrovel clearance guorontees there AN4-11 Bolt (2 Req) AN310-t Nul (2 Req) will be no interference under flying condition\ where AN960-tl6 Washer (2 Req) the cobl"s ore slack (i.e., extreme cold, etc). If neces sary, loosen cable• to moke this che<:k, and then adjust AN380-2-2 Cotter (2 Req) ccbltn. 01 in.s.tructed in step 1. f. AN515-8Rl6 Screw (2 Req) AN365-83::? Nvt (2 Req) AN960-8l Washer (4 Req) G. AN23-lS Bolt (1 Req) AN310-3 Nul (1 Reql AN%0-lOL Washer {1 Req) AN380-2·2 Cotter (1 Req) l-45-52214 Spacer (1 R«q) NOTE, The elevator travel stops ore not odjustcble. H. AN210-tA-4DR7-J'-4 Pulley (1 Req} AN4-10 Bolr (1 Req) AN310-4 Nut Ten•ion (1 Reo) Cob!e 30 oovnds AN320-2-2 Co�er o' Re<;} Surfoc& rove1 I 30' (+ 2./-0") u 20" (: 2") dow: Figure 3-6. Conventional Elevator Control System [8] 21 Table 3-2. Servo Response [9] Control Displacement Rate Limit, Bandwidth, Maximum Specific Limit, deg Deg/sec Flat, (6 dB Force or Moment Down) Hz (lAS = 105) Pitch -30 70 5(10) 9.9 rad/sec2 +20 Roll ±20 70 5(10) 9.2 rad/sec2 Yaw ±20 70 5(10) 4.2 rad/sec2 Thrust 0.6 * .05 g Normal +30 110 2(3) l. lg *Limited by aircraft engine to a firstorder time constant of 0.25 seconds. l The FBW elevator an d aileron systems incorporate redundant control channels. With the redundant channels, any substantial error between the commanded and actual control position is detected, and a switchover to a second servo is made. The fact that a channel has switched to the secondary servo is communicated to the pilot by a warning light. An "abort mode" disengagement system is installed which is activated pressing a disengagement thumb switch on the pilot's yoke. [9] A block diagram of the FBW rudder control system is presented in Figure 3-7. Inputs to the rudder are rudder position (Or), lateral accelerations (ay), and yaw ('If). Aileron, elevator, and throttle schematics are not presented in this document, but are of similar design as the rudder system. The location of the rudder hydraulic actuator is presented in Figure 3-8. One end of the actuator is connected to a bracket that is mounted to the fuselage, and the other end is connected directly to a control hom on the right side of the rudder. 22 Inputs �____...., Feedback Buffer Signal Null Amplifie Switches Amplifier IFoil ow and Servo 1--- and and up 1-- Sign f---+ Valve ain Pot! Summer g. Changer p pot By and pass !velocity [!r� Valve IXDCR � j Hyc ulic S o· Data Acquisiti on �r System Error L.. � Detector � u Conventional Rudder Pedals Figure 3-7. Fly-By-Wire Rudder Block Diagram 23 F(c.ucmr p,.�rt\o.r� A��a+cr EJ�or N � Ci�on� A�o�· Figure 3-8. Fly-By-Wire Rudder and Elevator Diagram The FBW elevator control system consists of two hydraulic servos, a primary and a secondary. During engagement of the system, a logic system allows fo r pressurization of the primary actuator fo r control inputs, and by-pass of the secondary actuator. If a failure of the primary actuator is sensed, the secondary actuator is pressurized and the primary actuator is placed in the by-pass mode. Should the logic system sense failure of both actuators, both are placed in the by-pass mode and the aircraft must be controlled using the conventional control system. The locations of the elevator control servos are presented in Figure 3-8. The primary actuator is located at the base of the empennage. This servo has a five-inch control hom mounted on the elevator and is connected to a bellcrank attached to the actuator. This control hom is connected to the underside of the elevator. The secondary hydraulic actuator is located just aft of the wing assembly at 173 inches aft of datum, housed within the fuselage. The secondary actuator has an eight inch control rod connected to a control hom. This control hom is connected to the conventional control system elevator cable. The FBW aileron control system consists of a primary actuator mounted on the underside of the right wing, and a secondary actuator mounted to the underside of the left wing (Figure 3-9). The operational priority system of the actuators is the same as stated previously fo r the elevator system. The actuators are connected to a bellcrank by a two inch rod. A nine-inch rod connects the bellcrank to the aileron surface bottom side. The available inputs to each of these controls are shown in Table 3-3. Independent control over heave is exercised through the Navion flap, modified to deflect up, as well as down through a ± 30 degree range. The upward motion provides increased lift modulation 25 Figure 3-9. Fly-By-Wire Aileron Control System 26 Table 3-3. Inputs of Moment Controls [9] CHANNEL INPUT FUNCTION VARIED Pitch Control stick displacement Control sensitivity Thrust lever Simulated moment due to thrust Column thumbwheel Simulated DLC moment Radar altitude Ground effect moment Airspeed Speed stability Angle of attack Static stability Pitch attitude Attitude hold sensitivity Pitch rate Pitch damping Flap angle Trim change from flap Flap rate Moment from flap rate Simulated turbulence Turbulence response Roll Lateral stick displacement Control sensitivity Sideslip Dihedral effect Roll rate Roll damping Yaw rate Roll due to yaw rate Rudder pedal displacement Roll due to rudder Simulated turbulence Turbulence response Yaw Rudder pedal displacement Control sensitivity Sideslip Directional stability Yaw rate Yaw damping Roll rate Yaw due to roll rate Wheel displacement Yaw due to aileron Simulated turbulence Turbulence response 27 authority and tends to minimize the problems of drag and angle of zero lift changes. Actuation is hydraulic, with a maximum available surface rate of 110 degrees per second. Inputs to normal force control presently available are shown in Table 3-4. Thrust and drag modulation is by a hydraulic servo on the engine throttle. The engine rpm is maintained by the propeller governor, which adjusts the blade pitch for the constant speed propeller. Inputs to the thrust/drag modulation system are shown in Table 3-5. Conventional Engine Control System For the conventional control system, the engine is controlled from the cabin by flexible individual push-pull controls (Figure 3-1 0). These push-pull controls are connected to the carburetor throttle lever, manual mixture and idle cut-off lever, and the carburetor air control lever on the bottom of the air mixing chamber. Modifications have been made to incorporate two hydraulic actuators in series for operation of the FBW system. When the FBW system is disengaged, these actuators operate as fixed links within the system. Operation of the actuators during FBW engagement is explained in the subsequent FBW paragraph. A six and twelve-inch push-pull rods are used to incorporate the actuators within the throttle system. Because N66UT is equipped with a Hartzell propeller, an additional flexible control is installed for regulating the propeller governor. Control knobs normally mounted on the cabin control panel have been replaced by lever style controls mounted on the right wall next to the copilot. The conventional throttle control system is presented in Figure 3-11, labeled number 1. 28 Table 3-4. Inputs to Normal Force Control [9] INPUT FUNCTION VARIED Control stick displacement Lift due to control (simulated elevator lift, or direct lift control integrated with column) Thrust lever displacement Lift due to thrust,direct lift control integrated with throttle Radar altitude Ground effect/lift; wind gradients Angle of attack Liftresponse to angle of attack, lift change at stall Simulated turbulence Turbulence response Table 3-5. Inputs to Thrust/Drag Modulation System [9] INPUT FUNCTION VARIED Control stick displacement Drag due to control (simulated control surface drag) Thrust lever displacement Thrust command/throttle sensitivity Radar altitude Ground effect drag changes; wind gradients Airspeed Drag change with speed Angle of attack Drag change with angleof attack Flap displacement Drag due to flap deflection 29 Pr:""or';;) Thr-.*'lt w....at'or Su..o�r':;S ·Throftl� A�mr Figure 3-10. Engine Throttle Control System 30 ------� v v Figure 3-1 1. N66UT Engine Controls and Instrumentation (1) Conventional, (2) Fly-By-Wire 31 Fly-by-Wire Engine Control System For the FBW control system, the engine is controlled through using the push-pull control system described in the conventional control section (Figure 3-1 0). The primary actuator controls the system while the secondary actuator acts as a fixed link. Should the logic system sense a failure of the primary actuator, the secondary actuator is pressurized fo r control input and the primary actuator becomes a fixed link. During failure of both actuators, both act as fixed links and the throttle must be controlled using the conventional control system. The FBW control levers are centrally located on the cabin control panel. The control panel is mounted with a hinge pin assembly, allowing the controls to be moved to gain access to computer variable adjustment knobs. When the system is engaged, the hydraulic actuators receive input signals from the analog computer based on movement of the throttle control. The FBW throttle control is presented in Figure 3-11, labeled number 2. INSTRUMENTATION. CALIBRATIONS AND CORRECTIONS The Navion is equipped with a standard array of aircraft instruments (Figure 3- 11). These include the manifold pressure, engine RPM, airspeed indicators, and altimeters. The manifold pressure and engine RPM gauges are connected to sensors in the engine. The airspeed and altitude indicators are connected to the ship's pitot-static pressure systems. Instrument calibration information was obtained from previous flight tests conducted at UTSI and is included in Appendix A, Figures 1-3. In addition to the standard instruments, the aircraft is equipped with a junction box (Figure 3-12) fo r access 32 (1) Figure 3-12. Fly-By-Wire Computer Interface to the variable-stability flight data computer described in the previous section. Data extraction instrumentation are labeled as fo llows: "1" is the junction box, "2" is the analog computer system, and "3" is the DaqBook.A static on-ground calibration was conducted to determine potentiometer dial settings that would provide equal control surface movement fo r equal control displacement between conventional and fly-by-wire system . Control movement distance measurements were taken fr om the conventional control system and correlated to the actual control surface response. The same measurements were made of the fly-by-wire control system, and potentiometers were 33 to match the conventional system deflection. This calibration resulted in the fo llowing dial settings fo r the axes of interest: • Aileron LBa = 58 Pedals NBr = 50 • Throttle XBt = 57 Elevator MBe = 68 Calibration data are included as Appendix A, Figures A-4 through A-19. Measurements required fo r handling quality analysis consisted of angular rates and attitudes, linear velocities and accelerations, and control positions. Accelerometers mounted on the test aircraft were located 98 inches aft of datum, 2.35 inches forward of the 100.35 inches aircraft center of gravity during the test. Vertically, the accelerometers were located at 109.0 inches above datum placing them 1.7 inches below the vertical C.G., located at 110.7 inches above datum. TEST METHOD AND PROCEDURES The short-term fr equency response flight-tests fo r the Ryan Navion consisted of three flights of 1.5, 1.9, and 0. 7 hours respectively. The flight tests were flown under daylight visual conditions at Tullahoma Municipal Airport, Tullahoma, Tennessee, on the 1 1h 9 h and 1 0 of October 1998 fo r data recording, and a flight on the 11th of November 1998, fo r validation of data. The essential test elements included 26 frequency sweeps of 120-second duration each and doublets at the frequency sweep condition. A summary of tests and test conditions are presented in Table 3-6. The flight tests were conducted at 90 knots calibrated airspeed (KCAS), and a density altitude (Ho) of 5000 fe et for comparison by UTSI, to wind tunnel data. The tests were flown within the existing envelope of the aircraft. The flight crew consisted of two crewmembers. Average gross 34 weight and CG for the first flight were 3,177 pounds and 100.37 inches aft of datum, with an average M.A.C. value of 24.45 %. Average gross weight and CG for the second flight were 3,199 pounds and 100.32 inches aft of datum, with an average M.A.C. value of 24.37 %. Fuel at engine start was fu ll at 39.5 gallons. Weight and Center of Gravity calculations are presented in Figure 3-13. The tests were conducted with the canopy closed, heater and carburetor heat off. Table 3-6. Tests and Test Conditions Test Axis Control Airspeed Altitude Average Average Average Station KCAS Ft. Ho OAT Gross e.G. oc Wt. (lbs) (Inches) Forward Pitch, Flight Frequency Roll, P, C!P 90 5000 4, 10 * 3188 100.35 Sweep Yaw, Throttle Doublet Pitch, Roll, P, C!P 90 5000 4, 10 * 3188 100.35 Yaw, Throttle * Indicates average temperature during the fly-by-wire flight test 35 Navion Weight and Balance N66UT Flight Identification: Conventional Freguency Sweep Crew Pilot: Dr. William Lewis Date: 9 Oct 98 Copilot: Charles Catterall TakeoffConditions Weight (lbs) Arm (in) Moment (in-lbs) Basic Aircraft 2546 100.79 256,605.69 Pilot/Copilot* 430 96.0 41,280 Laptop Computer 10 96.0 960 DaqBook Shelf 20 116.0 2,320 Fuel (39.5 gallons) 237 103.0 24.411 Total 242 325,576.69 CG = Moment/weight = 00.39@ i I CG Location 100.39 in. -It 83.66 in� = 16.73 in/68.35 in x 100= 4.5% M.A. Landing Conditions Weight Qbs) Arm (in) !2 MomentCj (in-lbs) Fuel Burned(2 1 .7gal) -130.2 103.0 -13,410.6 Total 312166.09 CG =Moment/weight = 100.32 i CG Location 100.32 in. - 83.66 in = 16.66 in/68.35 in x 100= 4.4% M.A. !2 Cj Flight Identification: FBW Freguency Sweep Crew Pilot: Dr. William Lewis Date: 10 Oct 98 Copilot: RandyL. Bolding TakeoffConditions Weight (lbs) Arm (in) Moment (in-lbs) Basic Aircraft 2546 100.79 256,605.69 Pilot/Copilot* 460 96.0 44,160 Laptop Computer 10 96.0 960 DaqBook Shelf 20 116.0 2,320 Fuel (39.5 gallons) 237 103.0 24,411 Total 272 328,456.69 CG = Moment/weight = 100.38@ i I CG Location 100.38 in. -l 83.66 in� = 16.72 in/68.35 in x 100= 4.46% M.A. Landing Conditions Weight Qbs) Arm (in) !2 Moment Cj(in-lbs) Fuel Burned (24.5gal) -147 103.0 -15,141 Total 125 313,31 5.69 CG = Moment/weight = 00.26@ i I CG Location 100.26 in. -It 83.66 in� = 16.6 in/68.35 in x 100= 4.29% M.A. Figure 3-13. Flight Test Weight and BalanceInform ation 36 The flight test technique to acquire data was to make a "frequency sweep" using the desired control. The sweep was performed by making control displacements to either side of trim at frequencies from 0.1 Hz (1 0 sec period) to 2.0 Hz (0.5 sec period). The control displacements were large enough to effect a noticeable aircraft response, but not so large as to generate large airspeed changes (+/- 10 knots) or translations. A desirable sweep contains at least three seconds of trim fo llowed by gradually increasing frequencies throughout the range of interest, and ending in three seconds of trim. The frequency range of interest was between 0. 1 ro8w and 2ro 180. Uniformity in the magnitude of control input throughout the frequencies is of no particular concern. It is both natural and acceptable fo r the magnitude at low frequencies to be quite small so that the attitude change will not be excessive, become larger in the mid-frequency range, then become smaller at high frequencies because ofthe physiological constraints of making rapid hand movements. Care must be taken to prevent unwanted movement of another control at the same frequency during the frequency sweep. While exact frequency inputs are not required, it is necessary to sweep the entire range of fr equencies. This is not an intuitive process and requires practice to preclude wasted flight test time. The ideal method is to practice in the aircraft and record the data. Data analysis should be used to identify any unexcited frequencies. Input and response data were recorded over a 120-second period (minimum to maximum frequency). Three frequency sweeps per axis were recorded fo r data reduction. Each frequency sweep ended with three seconds at trim condition fo llowed by a doublet control input to be used fo r time domain verification of results. The pilot not on 37 the controls held the laptop computer, activated the recording process, and counted seconds fo r control input. There is a high potential fo r structural damage to the aircraft during bandwidth testing, particularly if inputs are made with higher magnitudes or frequencies than required. Safety precautions were taken to start frequency sweeps with small magnitude inputs at lower frequency. Magnitude of input was increased through the middle frequency ranges and decreased again approaching higher frequencies. While performing the fr equency sweep tests, data were collected fo r handling quality analysis, fo r comparison of conventional and fly-by-wire control systems, and fo r a math model verification effort. SPECIAL PRECAUTIONS The fo llowing special precautions were observed: • All normal limits and emergency procedures contained in the aircraft operator's manual were reviewed prior to flight and observed during flight. • Crew coordination, test techniques, data recording and lookout responsibilities were reviewed prior to flight. • Incremental build-up techniques were utilized. DATA COLLECTION A personal computer based data acquisition system designed and marketed by IoTech Corporation, Cleveland, Ohio, was used for data collection (Figure 3- 14). The ® IoTech DaqBook 120 is a portable data acquisition system designed to interface with a 38 Figure 3- 14. IoTech Data Acquisition System [10] 39 ® notebook computer v1a the standard parallel port. DaqView , a spreadsheet style software provided with the system was used fo r data acquisition. System characteristics include 12-bit, 100-kHz AID conversion, 100 K reading/sec sampling and real-time storage to disk. Data acquisition fo r system identification required a 1 00 Hz sampling rate. The system allows 8 differential- or 16 single-ended inputs, expandable to 256 channels. Channel/gain sequencing is at 10 millisecond intervals. Programmable gain adjustments are available. Gain adjustments by a fa ctor of one, two, fo ur, or eight are included as selection options. DaqBook can support up to 800 Kbytes/sec total data transfer to a standard or enhanced parallel port interface or PC-Card link [ 1 0]. The conventional flight controls were not instrumented to provide data for control positions. Calibration data relating conventional control inputs to control surface deflection were used, and therefore an assumption of a rigid flight control system was made. The aircraft was not instrumented for continuous measurement of real time structural loads. The fo llowing data parameters were monitored and recorded fo r system identification: Control Positions • Pilot's yoke wheel (aileron) position (FBW) Channel 0 • Pilot's yoke (elevator) position (FBW) Channel l • Pilot's Pedal (rudder) Position (FBW) Channel 2 • Pilot's Throttle Position (FBW) Channel 3 Body Rates • Roll Rate (p) Channel 4 • Pitch Rate ( q) Channel S • Yaw Rate (r) Channel 6 40 Accelerations • ax Channel 7 • ay Channel S • az Channel 9 Actuators • Aileron Position Channel tO • Elevator position Channel l! • Rudder position Channel 12 • Throttle position Channe1 13 Attitudes • Roll Channel 14 • Pitch Channe1 15 41 Chapter IV DATA REDUCTION TECHNIQUE Daq View files were saved as binary files saved during the flight test. Conversion of binary to ASCII character format is completed using DaqView software. Flight data fi les were imported into a Microsoft Excel spreadsheet to convert data for each channel from volt readings to the respective engineering units using appropriate calibrations. Calibration data for recorded channels are presented as Appendix A, Figures 4 - 20. Files were saved as tab delimited ASCII files and transferred via network transfer (File Transfer Protocol) to a Silicon Graphics Incorporated Octane II parallel processing computer, using an IRIX 6.4 operating system. Time history data were processed into a CIFER usable format by a compiling program called Convert. The Convert program separated the fi le containing all sixteen channels into sixteen individual program files. The analysis of the flight-test data was conducted by first identifying frequency response from time histories and second by determining bandwidth and phase-delay from frequency responses. Requirements for bandwidth (Wsw) and phase delay ('tp) are as ® depicted in Figure 2-2. Data were analyzed using the CIFER 3.0 (£omprehensive Identification from .E(�quency Responses) program for aircraft system identification. CIFER® was developed at the NASA Ames Research Center by Mark Tischler and Mavis Cauffman to provide a tool for system identification. System identification is the determination of a mathematical description of aircraft dynamic behavior from measured aircraft motion. This process creates transfer functions from flight test data. The AFDD/NASA frequency-domain system identification procedure is shown in Figure 4-1. 42 Data Compatibility Multi·variable Frequency f--+ Aircraft r-+ Spectral Sweep Inputs & ...... , t-----.., State Estimation Analysis Conditioned Transfer-Function ,.....------� �------� Frequency-Responses Modeling & PartialCoherences + Freq.·Response Identification � Identification U Algorithm ------� Criterion � _ Mathematical Model Initial Values ··• Stability and Control Derivatives �---..., and Time Delays I �------' SensitivityAnalysis & Dissimilar flight Model Structure Verification Determination data not used in �---� identification '------l: APPLICATIONS:FCS design, Handling-Qualities, Simulation validation , __.....J Figure 4-1. Frequency Response Method fo r System Identification [9] 43 The frequency domain analysis is based on bandwidth and phase delay criteria (Bode plot). The bandwidth criteria are established using adequate phase and gain margin to insure piloted loop closure does not threaten stability. The foundation of the AFDD/NASA approach to data analysis is the extraction of a complete multi-input/multi output (MIMO) frequency-response database using an advanced multivariable spectral analysis with the Chirp-Z (advanced Fast Fourier) transform and composite optimal window technique [6]. The contaminating effects of any off axis control are removed by an inversion of the frequency response for all inputs to a single output. [11] Three steps were used to generate the frequency response database. The first step was to produce the single-input/single-output (SISO) frequency response from the time histories using the Chirp-Z transform. The second step was to condition the responses to account for the effect of secondary inputs. These conditioned multi-input/single-output (MISO) responses were the same as the SISO frequency responses that would have been obtained had no correlated controls been present during the frequency sweep of a single control. Step three was to combine multiple window lengths into a composite response. The overall result of these three steps (FRESPID, MISOSA, and COMPOSITE) was the rapid identification of a set of broadband frequency responses for all input/output pairs [5]. Data were analyzed first using the Frequency Response Identification Program (FRESPID),an integrated part of the CIFER. The four plots from FRESPID that are of interest are the time history, auto spectrum, coherence, and transfer function (Bode plot). It has been determined that better results can be obtained by applying Laplacian 44 integration to rate data rather than by using attitude data directly. The time history used two separate frequency-sweeps concatenated to make a single data sampling. As an example, the longitudinal stick inputs for two concatenated frequency sweeps, each lasting about 90 seconds is shown in Figure 4-2. No other stick input time histories are included. The inputs do not exceed 10 % and are smaller for low frequencies. Output data presented are pitch rates ( q). A desirable sweep contained several seconds of trim followed by gradually increasing frequencies throughout the range of interest. The o. frequency range of interest was between 0.1 ffiBw and 2.0 ro1s Uniformity in the magnitude of control input throughout the frequencies has no effect on the outcome. Irregularities in the shape of the input are usually interpreted as high-frequency data by the FRESPIDanalysis, and are acceptable. The input and output auto spectrum show the frequency distribution of the sweeps. As an example, Figure 4-3 presents the composite input auto spectrum for the elevator-surface input. This data indicates that good excitation was achieved over the range of 0.1 to 10.0 rad/sec. A drop-off in the auto spectrum occurs for input frequencies outside this range. No other auto spectrum data are included in this document. Relatively constant input power is desired throughout the frequency range of interest. A drop-off at higher frequencies is to be expected in the output spectrum. 45 �ILTERED INPUT TIME HISTORY �ILTERCD �U -PLT TIME HISTORY � ·�-----�------� .. oe ue 1i!0 ... ,.. ... TIME Figure 4-2. Time History: Longitudinal Stick Input to Pitch Rate Response 46 CCMF'0SITE INPUT AUTO SPECTRlJM FREQJE�CYCR�D/S[Cl Figure 4-3. Conventional Controls Longitudinal Input to Pitch Attitude Response Auto Spectrum The coherence function is a measure of the extent to which the output appears to be linearly related to the input. A value of one is ideal. For the purpose ofMIL-HDBK- 1797 data interpretation, a coherence function greater than 0.8 that does not oscillate is acceptable. Common sources of reduced coherence are: atmospheric turbulence, off-axis inputs at or near the evaluated frequency, sensor noise, insufficient excitation of the aircraft, or significant non-linearity in the relationship between control input and aircraft response. 47 The Bode plot presents the attitude response to the control input in terms of magnitude and phase versus frequency. An example plot is presented in Figure 4-4. The Bode plot from each axis was used to determine MIL-HDBK- 1 797 compliance in terms of bandwidth and phase delay. Bandwidth is generally a measure of the ability of a system output to satisfactorily follow input over a range of frequencies; good systems having high bandwidth and poor systems having low bandwidth. The system bandwidth (ro8w) is defined as the lesser of two frequencies from the Bode plot: ffioM, the frequency for 6dB of gain margin before neutral stability (180 degree phase), ro1 80; and ro135, the frequency for 45 degrees phase margin before the same point. Techniques used for calculating the gain margins are depicted in Figure 4-5. The phase delay ('tp) is a measure of the steepness with which the phase drops off after -180 degrees and indicates the behavior of the aircraft as the pilot attempts to control the aircraft with inputs at frequencies higher than the bandwidth frequency. Aircraft with large phase delays tend to be prone to pilot induced oscillation (PIO). If the phase between ro180 and 2 ro180 is linear, two points may be used to define 'tp. If it is non linear, a linear curve fit must be applied. The recommended procedure for performing a curve fit of the data is to use a linear, least squares fit. This will eliminate any uncertainty in the determination of linearity, and assure that specification compliance numbers will be similar regardless of individual judgement [12]. 48 � 0 s w• Q"' w '!��----��··----��·�·----��--� "1 w u z: w.�· w• � ! u ! . I FREQUENCY !RAOtSECl STLONSWP_FRE_00000_LONS_Q PUi: 360.00 Pw FWLONSWP_FRE_000D0_LONS_Q Ph: : -1 Figure 4-4. Typical CIFER Bode Plot; A Comparison of the Fly-by-wire and Conventional Control Systems 49 Figure 4-4 contains CIFER data plot coding which can be located at the top or bottom of the data plot, depending on which utility was used to process the Bode Plot. The fo llowing summation is presented to explain what each part of the coding system means to the reader. For the Heading : STLONSWP_ FRE_OOODO_LONG_Q The first two letters in the label "ST" indicate the control system. In this case it is the standard (conventional) flight controls. The second part of the label "LONSWP" is the case name assigned fo r CIFER identification. The third section of the label "FRE" is the CIFER program used fo r data analysis, in this example, FRESPID. "OOODO" was assigned to indicate a 40-second window. Each position has a window size assigned fo r the program to process. During use of COMPOSITE, several windows could be used simultaneously fo r better identification. Within a window identifier of ABCDO, A = 40 sec, B = 30 sec, C = 20 sec, and D = 10 sec window lengths. "LONG" indicates the input fo r the data. "Q" (pitch rate) indicates the channel from which the data were recorded. A shortcoming in the utility used fo r individual control system data plots neglects to indicate when a laplacian integration has been used. No provisions are available to change source information labels. In the case above, an appropriate symbol in place of "Q" would have been "8" fo r pitch attitude. This problem is apparent with many Bode plots included within Appendix B. Particular attention must be paid to the figure title to insure an understanding ofdata being presented. "Ph: 360" indicates any phase shift required for data analysis. "Pw: - 1 " is the power of a Laplacian integration 50 (s) multiplied into the transfer fu nction. Above, an s-1 has been multiplied into the transfer fu nction for pitch attitude acceleration to obtain pitch attitude change. 51 Phase Delay : phase is between "Note: if nonlinear WJBO and 2w;so , 'P ·snail be de termined fr om o linear le ast squares fit to phase curve between w180 and 2w1so Rote Response -Types: is lesser w6w of wawgain ond w ewphcse Atti tude Commend/Attitude Hold Response-Types (ACAH): waw : wawphose H 22 a g n i lll t u wew�cin d e 0 1:i las d B (X : 8,¢,\f!} l - (Xi =F5or o5) l :l I -2ii -9 p 8 h a 5 e cp -182 (de g) Frequency (rod /sec l {log scale l Figure 4-5. Definitions ofBandwidth and Phase Delay [6] 52 Chapter V DISCUSSION OF TEST RESULTS DATA ANALYSIS CIFER programs FRESPID, MISOSA, and COMPOSITE were used to evaluate frequency-response data. Corrections to FRESPID data for off-axis response were minimal. Final COMPOSITE response data plots are presented in Appendix B. Frequency response identification was completed using the angular rate variables (p, q, r) rather than the an gular attitude variables (<)>, 8, 'JI) because the middle and high frequency content of the rate variables is greater. The middle to high frequency signals are better suited for identification of the bandwidth and phase-delay parameters. When the identification of the low-frequency characteristics is more important, the attitude response variables are better suited for the analysis. The COMPOSITE program determines the required attitude responses from the rate responses by applying a Laplacian iteration to the magnitude and phase curves. The bandwidth and phase-delay parameters needed for demonstration of specification compliance are defined in terms of the attitude frequency response in Figure 2-2. As stated previously, the bandwidth, c:osw, for a rate-response type system, like the Navion, is the lower of two frequencies: one, based on a gain margin of 6 dB; and the other, based on a phase margin of 45 ffiGM, rom, degrees. Bandwidth calculations for all data plots within this document resulted in a phase margin of 45 degrees (ro135) determining bandwidth limits. No minus 6 dB gain margins were reached during this test. 53 GENERAL Frequency sweep data are presented from a frequency of 0.6 rad/sec and truncated with a 5 Hz filter. Pitch, roll, and yaw control on axis responses had Level 1; Category C ratings, and Level 2; Category A ratings. Figure B-2 was used to demonstrate calculation techniques used in determining time delays from doublet time histories. Considering the high rate at which the initial doublet input is made, calculation of time delays from frequency-response data were taken at 20 rad/sec for comparison. Doublet control input information for all FBW systems were contaminated by interference, making this data unusable. Time delay differences between the conventional and FBW systems were calculated at FBW bandwidths rom. During the handling qualities data validation flight, a technique was used to identify the maximum amplitude response by a control sweep of increasing frequency. Once identified, this amplitude was maintained fo r ten seconds and the number of cycles noted. An approximate frequency was calculated. Validation flight damping estimates were determined by displacing the aircraft away from trim and by observing the number of overshoots until returning to trim. Approximation of damping was accomplished using the fo llowing fo rmula: 7-(number of overshoots) 10 Flight check of spiral mode during the validation flight indicated positive stability with an 11-second time-to-half amplitude in the left and right roll directions. 54 LONGITUDINAL CONTROL INPUT Pitch Mode, Conventional Control System The difficulty in generating large low-frequency input signals in the pitch axis resulted in poor pitch response identification at the lowest input frequencies. However, the coherence function indicates good identification in the range of 0.6 to 10.0 rad/sec, which is satisfactory for specification compliance. Figure B-1 indicates a heavily damped 1st order pitch mode at 3.2 rad/sec. The resultant time constant is 313 milliseconds (ms). A heavily damped 1st order system was evident to the pilot during the fl ight. During the verification flight 6 pitch attitude cycles were experienced in 10 seconds, equating to a period of approximately 3.78 rad/sec. No overshoot of pitch attitude was noticed during the flight. Bandwidth from frequency data was at 4.84 rad/sec. The effective time delay is 487 ms measured at bandwidth (ro135). Time delays of 487 ms at bandwidth will be very noticeable to the pilot as sluggishness in the control response. During the flight, attitude changes to control inputs decreased as frequency of movement increased. Since the phase curve is nearly flat at co180, the phase-delay for the conventional control pitch response is negligible. Figure 5-1 presents time history data for the conventional control system at bandwidth. Control deflection and pitch attitude change are presented. A line is drawn at the maximum control displacement, and another line is drawn at the point of maximum nose down deflection. Time delay indicated by the time history was determined to be 480 ms, only an 8 ms difference from the frequency determined time delay, and thus verifying the calculation. 55 8 7 +------� E ESE 4 �------�r------�----�------N--� £� c C'::S ...c Pitch Attitude Response (deg) u 3 Nose Down 2 +------4------4 I I I Longitudinal Control I I I Displacement (in) +- 480 rns --+l I I 70 71 72 Time (sec) Figure 5-l. Longitudinal Time History (Conventional Controls); Longitudinal Input to Response 56 Pitch Mode, Conventional Control System Doublet Figure B-2 presents the doublet input used m validating test results for the conventional control system. The time delay indicated fo r the doublet initial input was approximately 172 ms. The frequency derived time delay was 161 ms, only 12 ms diffe rence from the doublet time delay, validating the test data. Handling-qualities experience indicates that the pilot may notice an equivalent time delay in excess of 180 ms. Initial aircraft response to step-type input should be within 172 ms. No delay was noticeable to the pilot during the flight. Pitch Mode, FBW Control System Figure B-3 data indicated a heavily damped 1st order pitch mode at 3.68 rad/sec fo r the FBW control system. Bandwidth was at 3.68 rad/sec. The time constant calculation resulted in a value of 278 ms. Since the phase curve is almost flat near ro 1so, the phase-delay calculation fo r the FBW control pitch response was calculated using a least-squares curve fitroutine, resulting in 'tp = 27 ms. The effective time delay measured at ro135 was 655 ms. A time delay of 655 ms at bandwidth will be very noticeable to the pilot. During the flight, sluggishness in control response was noticeable as frequency of control movement increased. At high frequency control inputs, a state was reached of no attitude change with control input. The time delay indicated by the frequency data is approximately 192 ms. Initial aircraft response to step-type input should be within 192 ms. Any delay in aircraft response was unnoticeable to the pilot. 57 Pitch Mode;Conventional and FBW Control Systems Figure B-4 presents both conventional and FBW system responses relating longitudinal stick to pitch attitude. The FBW control system closely models conventional control system responses. A time delay difference of 9.5 ms between FBW and conventional control systems was measured at FBW bandwidth of 3.68 rad/sec. This small difference in time delay is not noticeable to pilots. FBW system electrical component delays and actuator dynamics did not affect phase delays until beyond bandwidth at approximately 10 rad/sec. This value correlates directly with the 9.9 deg/s2 actuator maximum specific fo rce response limit specified in Table 2-2. For input frequencies below 0.6 rad/sec and above 10 rad/sec, the coherence function becomes erratic and the transfer-function identification is not reliable. Airspeed variation during longitudinal testing was ± 5 knots. Close monitoring was required to prevent control migration forward from the original trim point. Any deviation from the original trim point would result in a change to trim airspeed. Pitch Mode to Normal Acceleration The Bode plot of normal acceleration response to longitudinal control input for both control systems is shown in Figure B-5. The magnitude curve is flat between 0.6 and 3 rad/sec, indicating a normal acceleration response to control input. The pilot would perceive an upward force of the seat as the nose pitches up, and a downward drop in the seat as the nose is pitched down. The phase plot indicated approximately a 90-degree phase shift in the response. The pilot would perceive a reduction in the normal acceleration with an increase in the frequency of control movement. These flight 58 characteristics were verifiedby the pilot qualitative assessment. The coherence function fo r the normal acceleration response is strong over the frequency range of 0.6 to 3 rad/sec, with the drop-off above 4.5 rad/sec. Pitch Mode, Cross-coupling Figure B-6 presents an example of the cross-coupling off-axis responses related to longitudinal inputs. Off-axis coherence plots indicated no linear correlation between inputs and outputs. To the pilot, this equates to minimal unwanted roll or yaw coupled response being generated with a commanded pitch attitude change. Aircraft behavior during the flight test indicated no unwanted off axis responses associated with longitudinal control inputs. LATERAL CONTROL INPUT TO ROLL ATTITUDE CHANGE Roll Mode, Conventional Controls The Bode plot fo r roll attitude response to control input for the conventional control system is presented in Figure B-7. The response is characterized by a heavily damped 1st order roll mode at 2.5 rad/sec. The time constant calculation resulted in a value of 400 ms. Bandwidth was determined to be 5.41 rad/sec. Pilot assessment verifiedthe data. During the data verification flight, a maximum output to input response frequency was visually established using the conventional control system. Results were 5.5 roll attitude cycles in 10 seconds, equating to a period of approximately 3.45 rad/sec. No overshoot was experienced with roll mode responses. The effective time delay was 436 JlS measured at ro135. A time delay of 436 ms at bandwidth will be noticeable to the pilot. During the frequency sweep, sluggishness in control response was noticeable as 59 frequency of control movement increased. A state of no roll attitude change with control input resulted at a higher frequency control movement. Since the phase curve is nearly flat near ro 18o, the phase-delay calculation fo r the conventional control roll response is negligible. Roll Mode, Conventional Control Doublet Figure B-8 presents the doublet input used to validate test results for the conventional control system. The time delay indicated fo r the doublet initial input is approximately 154 ms. Calculation from frequency-response data at 20 rad/sec resulted in a delay of 147 ms, a 7 ms difference from the doublet time delay, validating the test data. Initial aircraft response to a step input should be within 154 ms. This delay was unnoticeable during the flight. No roll attitude overshoot or oscillations were experienced during the test. Roll Mode, FBW Control System The Bode plot for roll attitude response to control input for the FBW control system is presented in Figure B-9. The response is characterized by a heavily damped 1 st order roll mode at 2.5 rad/sec. The time constant calculation resulted in a value of 400 ms. Bandwidth frequency was 3.85 rad/sec. The effective time delay was 612 ms measured at ro 135• A time delay of 612 ms at bandwidth will be very noticeable to the pilot. During the flight, delay in control response was noticeable as frequency of control movement increased. A state of no roll attitude change with control input resulted at highest control movement frequenciestested. Using a least-squares curve fit routine, the phase-delay calculation for the FBW control roll response was "tp = 30 ms. 60 The time delay indicated by the frequency data at 20 rad/sec was approximately 146 ms. Any delay in control response was unnoticed by the pilot during the flight. Roll Mode, Conventional and FBW Control Systems Figure B-1 0 presents both conventional and FBW system responses relating lateral stick input to roll attitude response. The coherence function for the yaw response in both systems was unreliable below 0.9 rad/sec, and dropped below a value of 0.8 at frequencies above 10.8 rad/sec. The FBW control system closely models conventional control system responses, with a time delay difference of 8.9 ms, measured FBW roll bandwidth 3.85 rad/sec. This small difference in time delay will not be noticeable to the pilot. FBW system electrical component delays and actuator dynamics do not begin to affect phase delay until approximately 6 rad/sec, which is beyond bandwidth for this system. Table 3-2 indicates a roll actuator limit for maximum specific force or moment of 9.2 rad/sec2. For input frequencies below 0.6 rad/sec and above 17 rad/sec, the coherence function becomes erratic and the transfer-functionidenti fication is not reliable. Adverse Yaw The Bode plot for yaw attitude response to roll control input fo r both the conventional and FBW control systems is presented in Figure B-1 1. This adverse yaw response is characterized by a moderately damped 2nd order mode with a natural frequency of about 2.4 rad/sec, and a damping ratio of approximately 0.5. This mode manifests itself at 180 degrees out of phase response to control input, and creates an adverse yaw effect. As the aircraft is rolled right, the nose of the aircraft yaws left. When rolled left, the aircraft yaws to the right. During roll axis frequency sweeps, a 61 continuous yawing motion was noted. This was apparent to the pilot during flight testing. All other cross-coupling responses showed no linear correlationwith roll inputs. Roll Mode Lateral Acceleration (AY) The Bode plot of the Y -axis acceleration response to lateral control input for both control systems is shown in Figure B-12. The magnitude curve rises slightly between 0.6 and 2.3 rad/sec and drops off rapidly above that frequency. The characteristic of the magnitude data indicated an increasing lateral acceleration response to control input, with a drop-off in response fo r fr equencies above 2.3 rad/sec. The phase plot indicated a negative Y-axis acceleration with positive roll input. This was perceived as a sideward fo rce of the aircraft away from the direction ofroll as the pilot maneuvers in the roll axis. This is opposite of what was experienced by the pilot during this flight. This was due to the accelerometers being located below the vertical CG of the aircraft. The phase plot also indicates approximately a 90-degree phase shift in the response before coherence drops off. This data indicates the pilot would experience a reduction in the sideward fo rce with an increase in the fr equency of control movement. This was noticed during the flight. The coherence function for the y-axis acceleration response within acceptable range over the frequency range of 0.6 to 2.3 rad/sec, with a drop-off above 2.3 rad/sec. 62 PEDAL CONTROL INPUTTO Y AW ATTITUDE CHANGE Yaw Mode, Conventional Controls The Bode plot for yaw attitude response to control input for the conventional control system is presented in Figure B-13. The response is characterized by a lightly damped 2"d order mode with a natural frequency of approximately 2.5 rad/sec, and a small damping ratio. Flight observations indicated a lightly damped oscillatory response. Bandwidth at ro135 was 2.9 rad/sec. During the verification flight 4.5 yaw attitude cycles were experienced in 10 seconds, equating to a period of approximately 2.8 rad/sec. The yaw response had 5 overshoots of trim before recovery, indicating a damping ratio of approximately 0.2. The effective time delay was 812 ms measured at ro135. A time delay of 812 ms will be very noticeable to the pilot in all tasks associated with yaw axis control near 2.9 rad/sec. During the flight, yaw response became sluggish as frequency was increased until a frequency of no response to control input at approximately 23 rad/sec. The pilot experienced slight overshoot and oscillation (about the desired attitude). Small in-flight oscillations in the yaw axis were noticed with no associated control inputs, indicating weak directional stability about the trim point. Since the phase curve is nearly flat at ro180, the phase-delay for the conventional control yaw response is negligible. Data for sideslip (\jl) were not recorded, therefore no doublet information is available for the yaw axis. The time delay indicated by the frequency data at 20 rad/sec was approximately 157 ms. Delay in initial aircraft response to control input was not noticeable. Figure 5-2 presents time history data for the conventional yaw control system at bandwidth. Due to lack of instrumentationfor the yaw attitude, control deflections are 63 0.6 E ·;::: 0.5 f-< � 01J :::: � ..c: u 0.4 0.3 0.2 +------+�------�I I I I 700 ms Lateral Acceleration (g's) I I I I �I I� 0. 1 -0.1 �� : 48 49 50 51 52 Time (sec) Figure 5-2. Conventional Control Pedal Input to Lateral Acceleration Time History 64 plotted against lateral accelerations, which closely represent a sideslip attitude. Time delay indicated by the time history was determined to be approximately 700 ms, 112 ms difference from the frequency determined time delay. Phase delay differences between sideslip and lateral acceleration probably accounts for some of the difference. Yaw Mode, FBW Control System The Bode plot fo r yaw attitude response to control input for the FBW control system is presented in Figure B-14. The response is characterized by a lightly damped 2nd order mode with a natural frequencyof about 2.4 rad/sec, and a small damping ratio. Bandwidth at ro135 was 2.83 rad/sec. The effective time delay was 833 ms measured at ro135• A time delay of 833 ms at bandwidth will be very noticeable to the pilot in all tasks associated with yaw axis control. During the flight, yaw response became sluggish as frequency was increased until a point of no response to input at 18 rad/sec. The phase delay calculation for the FBW control yaw response was 'tp = 29 ms. Data fo r sideslip (\ll) were not recorded, therefore no doublet information is available for the yaw axis. The time delay indicated by the frequency data at 20 rad/sec was approximately 196 ms. The pilot noticed no delay in aircraftyaw to pedal input. Small cockpit design combined with pilot size resulted in limited space for control movement. During the pedal sweep of the FBW control system, the copilot was inadvertently making aileron inputs at the same frequency by the contact of his knees with the yoke assembly. An additional data run was required because of problems encountered with frequency-sweep techniques. This problem can be avoided by not resting your hands on your knees during input. Alternatively, practicing the input with 65 smaller magnitude may solve the coupled input problem. In worst case, a smaller pilot may be required. Yaw Mode, Conventional and FBW Control Systems Figure B-17 presents both conventional and FBW system responses fo r pedal control input-to-yaw attitude response. The FBW control system closely models conventional control system responses. There was no measurable time delay between the systems at FBW bandwidth, 2.4 radlsec. FBW system electrical component delays and actuator dynamics do not begin to affect phase delays until a frequency of approximately 5 radlsec, beyond bandwidth fo r this system. Table 3-2 indicates a yaw actuator limit fo r maximum specific force or moment of 4.2 radlsec2• For input frequencies below 0.9 rad/sec and above 17 radlsec, the coherence function becomes erratic and the transfer function identification is not reliable. Dihedral Effect The Bode plot fo r roll attitude response to pedal control input fo r both the conventional and FBW control systems is presented in Figure B-16. The dihedral response is characterized by a moderately damped 1 st order mode at 2. 7 radlsec. The resultant time constant is 370 ms. This response manifests itself at 90 degrees out of phase with the control input, and creates a positive dihedral response. During flight positive sideslip resulted in a negative roll. As the aircraft was yawed nose right (left sideslip), the aircraft pitched slightly nose down androlled to the right, away from the sideslip. Nose down pitch rate was slightly faster with a right sideslip. 66 Yaw Mode Lateral Acceleration (A Y) The Bode plot of the Y -axis acceleration response to pedal control input for both control systems is shown in Figure B-17. The magnitude curve rises between 0.6 to 2.5 rad/sec and drops off rapidly at higher frequencies. This indicates an increasing acceleration response to control input, with a drop-off in response for frequencies beyond 2.5 rad/sec. The phase plot indicates an in-phase Y-axis acceleration up through 2.0 rad/sec, with a 180-degree phase shift for frequencies above 4.0 rad/sec. This is perceived as a sideward force of the aircraft into the direction of yaw. This result was verified by the pilot during flight. All other cross-coupling responses indicated no linear correlation with yaw inputs. THROTTLE CONTROL INPUTTEST RESULTS The Bode plot of the X-axis acceleration response to throttle control input for both control systems is shown in Figure B-18. Test result coherence for all channels recorded indicated poor results. Transfer-function identification and handling qualities evaluation for all throttle inputs were inconclusive. Table 3-2 indicated a throttle servo response bandwidth of only 0.6 rad/sec. Inputs of 0.6 rad/sec were at the lower bound of the test data interval. Throttle sensitivity was a concern during testing. The FBWthrottle was more sensitive and less predictable in response than the conventional control. Frequency inputs to both systems were made to an estimated 5 percent either side of trim, however throttle response was inconsistent. Engine accelerations were large and non-linear, with a 67 sluggish response after initial movement followed by rapid acceleration. Responsiveness of the propeller was not sufficiently adequate to maintain consistent RPM. During flight 70- 100 RPM engine fluctuations were experienced. Control response was jerky, and the engine neared stall during one iteration indicating a need for a control fixture for this test. 68 Chapter VI CONCLUSIONS The purpose of this thesis was to evaluate the handling quality characteristics of the Ryan Navion (N66UT) using frequency-domain test techniques. Conventional and Fly-by-Wire flight control systems were compared to determine any time differences in response between the two control systems. Evaluation of aircraft handling qualities were compared to fixed wing flying quality specifications outlined in Mll..-HDBK-1797. Due to the lack of specifications regarding low gain roll and yaw maneuvers, specification compliance in these axes was assumed to follow pitch attitude specification data. The flight test developed and conducted by the University of Tennessee Space Institute resulted in the following conclusions: 1. Frequency-domain system-identification methods are well suited to aircraft flight-control and handling-qualities an alysis since many current design specifications, design and analysis techniques, and acceptance flight-test techniques are based in the frequency domain. Frequency-domain methods extract large amounts of broad frequency flight-test data using a minimal amount of flight-time. 69 2. The response characteristics of the Navion N66UT are stable for the longitudinal, lateral and rudder control inputs. All bandwidth frequencies were determined by 45 degree phase margins (w135). The handling quality parameters meet or exceed proposed Level 1, Category C requirements, and Level 2, Category A requirements for fixed-wing military aircraft. A summary table of test results is presented in Table 6- 1. 3. The fl y-by-wire system closely mirrored conventional control behavior well beyond bandwidth frequency. Results are presented in Table 6-2. Time history data from fl ight test were used to verify frequency time delay calculations at bandwidth for the conventional controls, longitudinal control input to pitch attitude and pedal control input to lateral acceleration. 4. Time delays in excess of 400 ms at bandwidth were experienced in pitch, roll and yaw control axis. This would prohibit use of the variable stability airplane, N66UT; to simulate any high-bandwidth, minimal control delay designed aircraft. 5. Activation of the FBW system caused interference with attitude response data channels 14 (roll attitude) and 15 (pitch attitude) signals, and prevented the use of doublet information for FBW verification. 70 Table 6-1. Summary of Test Results Control Input Output Bandwidth Phase Time Delay Time System Delay at Constant rons tp rom (rad/sec) (ms) (ms) (ms) Conventional Long stick Pitch(S) 4.84 Negligible 487 313 (ST) Lat stick Roll() 5.41 Negligible 436 400 nd Pedal Yaw(\jf) 2.9 Negligible 812 2 order Fl::z:-b�-wire Long stick Pitch(8) 3.68 27 655 278 (FW) Lat stick Roll() 3.85 30 612 400 nd Pedal Yaw(\jf) 2.83 29 833 2 order Table 6-2. Time Delay in the Fly-by-Wire Control System with Respect to Conventional Controls Longitudinal Lateral Directional (Pedal) Time Delay Difference* 9.5 ms 8.9 ms neglig_able * calculated at FBW control system bandwidth 71 6. The Navion demonstrated the following flightcharacteri stics: • A positive dihedral effect with sideslip input. • Adverse yaw response to roll attitude changes were noticed during flight and verified in the frequency domain data. 7. Control delays due to hydraulic actuator dynamics did not affect pitch, roll, or yaw FBW control response until beyond bandwidth frequencies. 8. Throttle response data was below acceptable levels of coherence (0.8). No conclusions could be made regarding specification compliance. Table 3-2 indicates a servo response bandwidth of only 0.6 rad/sec. Inputs of 0.6 rad/sec were at the lower bound of the test data interval. Control sensitivity was a concernduring testing of the conventional and FBW throttle systems. The FBW throttle system was more sensitive to control changes and less predictable in response than the conventional system. Frequency inputs were made to an established 5 percent either side of trim, however throttle response was inconsistent. Engine accelerations were large and non-linear, with an initially sluggish response to control movement followed by rapid acceleration. 9. Lateral acceleration data indicated the opposite of observed aircraft accelerations during roll inputs. This false lateral acceleration data is the result of accelerometer package location below the vertical CG. 72 I 0. An additional frequency sweep of the FBW roll control and pedals were required due to improper frequency sweep techniques. 11. Inputs were inadvertently made to roll controls during the pedal sweep of the FBW control system. This input was due to the pilot's knees contacting the yoke assembly and making roll inputs at the same frequency as the pedal inputs. An additional roll frequency sweep was made due to delays in input and excessive control displacement. 73 Chapter VII RECOMENDATIONS 1. Review the instrumentation and data collection equipment prior to further testing. Determine the cause for the loss of pitch (ch 15) and roll (ch 14) attitude data during use of the Fly-by-Wire control systems. 2. Conventional control systems need to be instrumented for accurate position data. Using an assumption of a rigid control system does not allow for determination of any control hysteresis. 3. Fixtures should be considered to control range of motion during frequency-domain testing to increase input consistency. At a minimum, a control fixture should be designed for throttle sweeps of both conventional and Fly-by-Wire control systems. 4. Insure that the pilot performing the frequency sweep has adequate space to avoid unwanted control movements. If the cockpit area is not large enough to facilitate the pilot and all of his equipment, consideration should be made to using smaller pilot or different flight equipment. 74 5. Increase pilot training in frequency-domain testing techniques to avoid inadvertent off-axis response inputs while sweeping frequencies in the tested axis. Practice of frequency-sweep techniques using the control system to be tested should be arranged. 6. Locate the accelerometer package on the vertical and longitudinal center of gravity. If this is not possible, corrections should be made to accelerometer data to account fo r distance between sensor and center of gravity. 75 REFERENCES 76 REFERENCES 1. Department of Defense Handbook, MIL-HDBK-1797, "Flying Qualities of Piloted Aircraft", 19 December 1997. 2. Technical Memorandum,NASA TM 110369, USAATCOM TR 95-A-007, "System IdentificationMethods fo r AircraftFlig ht Control Development and Validation", October 1995. 3. Cook, M.V., "Flight Dynamics Principles", New York, NY: John Wiley & Sons, Inc., 1997. 4. Hob, Roger H., Mitchell, D. G., Ashkenas, I. L., Klein, R. H., Heffley, R. K., and Hodgkinson, J., "Proposed MIL Standard and Handbook-Flying Qualities of Air Vehicles". AFWAL-TR-82-3081, vol. 2, 1982. 5. Ham, Johnnie A., Tischler, Mark B., "Flight Testing andFrequency Domain Analysis fo r Rotorcraft Handling Qualities Characteristics", AHS Specialists' Meeting on Piloting Vertical Flight Aircraft,20-23 Jan. 1993, San Francisco, California. 6. Tischler, Mark B., "Identification and Verification of Frequency-Domain Models for XV-15 Tilt-Rotor Aircraft Dynamics in Cruising Flight", AIAA Journal of Guidance, Control, and Dynamics,July 1986. 7. Military Specification, MIL-F-8785C, "Flying Qualities of Piloted Airplanes", 5 November 1980. 8. ANON., "Navion Service Manual", First Edition, revised January 1951. 9. ANON., "The Variable-Response Research Aircraft", PrincetonUniversity, no date. 10. IoTech Sales Literature, IoTech Industry, Cleveland, Ohio, 1997. 11. ANON., NASA Conference Publication, NASA CP 10149, USAATCOM TR-94-A- 017, "Comprehensive Identification fromFrequency Responses", Class Notes from short course held at NASA Ames Research Center, Moffett Field, California, September 19-23, 1994. 12. Lewis, W.D., "An Aeroelastic Model Structure Investigation for a Manned Real Time Rotorcraft Simulation," American Helicopter Society 48th Annual Forum, St. Louis, MO, May 1993. 77 REFERENCES (continued) 13. Abbott, William Y., "Engineering Evaluation of Aeronautical Design Standard (ADS)-33C, Handling Qualities Requirements for Military Rotorcraft, Utilizing an AH- 64A Apache Helicopter", AVSCOM Project No. 87-17, November 1991. 78 BIBLIOGRAPHY 79 BIBLIOGRAPHY Advisory Report, AGARD-AR-279, "Handling Qualities of Unstable Highly Augmented Aircraft", May 1991. Abbott, William Y., "Engineering Evaluation of Aeronautical Design Standard (ADS)- 33C, Handling Qualities Requirements for Military Rotorcraft, Utilizing an AH- 64A Apache Helicopter", AVSCOM Project No. 87-17, November 1991. Aeronautical Design Standard, ADS-33D, "Handling Qualities Requirements fo r Military Rotorcraft", July 1994. ANON., "Navion Service Manual", First Edition, revised January 1951. ANON., "The Variable-Response Research Aircraft", Princeton University, no date. Basile, P.S., Gripper, S.F., Kline, G.F., "An Experimental Investigation of the Lateral Dynamic Stability of a Navion Airplane", Princeton University, June 1970. Buxo, Arturo Manual, "Performance of the Variable-Stability Navion with Flap Deflection", MS thesis, The University ofTennessee Space Institute, May 1995. Conference Publication, NASA CP 10149, USAATCOM TR-94-A-017, "Comprehensive Identification fromFrequency Responses", Class Notes from short course held at NASA Ames Research Center, Moffett Field, California, September 19-23, 1994. Cook, M.V., "Flight Dynamics Principles", New York, NY:John Wiley & Sons, Inc., 1997. Fletcher, Jay W., "Obtaining Consistent Models of Helicopter Flight-Data Measurement Errors Using Kinematic-Compatibility and State-ReconstructionMethods", 46th Annual Forum of the American Helicopter Society, May 1990, Washington, DC. Flight Test Manual, USNTPS-FTM-No. 107, "Rotary Wing Stability andControl", Patuxent River, Maryland: U.S. Naval Test Pilot School, Naval Air Warfare Center. Ham, Johnnie A., Tischler, Mark B., "Flight Testing and Frequency Domain Analysis for Rotorcraft Handling Qualities Characteristics", AHS Specialists' Meeting on Piloting Vertical Flight Aircraft, 20-23 Jan. 1993, San Francisco,Calif ornia. Hoh, Roger H., Myers, Thomas T., Ashkenas, Irving L., Ringland,Robert F., Craig, Samuel J., AFWAL-TR-81-3027, "Development ofHandling Quality Criteria for Aircraftwith Independent Control of Six Degrees ofFreedom", April 1981. 80 BIBLIOGRAPHY (continued) Hoh, Roger H., Mitchell, D. G., Ashkenas, I. L., Klein, R. H., Heffley, R. K., and Hodgkinson, J., "Proposed MIL Standard and Handbook-Flying Qualities of Air Vehicles". AFWAL-TR -82-3081, vol. 2, 1982. Kimberlin, Phd, Ralph D., "Stability andControl Flight Testing Lecture Notes", Tullahoma, Tennessee: The University of Tennessee Space Institute. Lewis, W. D., "An Aeroelastic Model Structure Investigation for a Manned Real-Time Rotorcraft Simulation," American Helicopter Society 48th Annual Forum, St. Louis, MO, May 1993. Military Handbook, MIL-HDBK-1797, "Flying Qualities of Piloted Aircraft", 19 December 1997. Military Specification, MIL-F-8785C, "Flying Qualities ofPiloted Airplanes", 5 November 1980. Military Specification, MIL-F-83300, "Flying Qualities ofPiloted V/STOL Aircraft", 31 December 1970. Ockier, Carl J., Pausder, H.-Juergen, "Experiences with ADS-33 Helicopter Specification Testing and Contributions to RefinementResearch", AGARD Flight Mechanics Panel Symposium on Active Control Technology, Turin, Italy, May 1994. Pierson, Brett M., "Flight Test Safety: Lessons Learned from AN S-3B Flight Test Mishap During a Frequency Domain Test", MS thesis, The University of Tennessee Space Institute, August 1998. Technical Memorandum, NASA TM 89422, USAAVSCOM TM 87-A-1, "Demonstration ofFrequency-Sweep Testing Technique Using a Bell 214-ST Helicopter", April 1987. Technical Memorandum, NASA TM 110369, USAATCOM TR 95-A-007, "System IdentificationMethods for AircraftFlig ht Control Development and Validation", October 1995. Technical Note, NASA TN D-5857, "Full-Scale Wind-Tunnel Investigation of the Static Longitudinal and Lateral Characteristics of a Light Single-Engine Low-Wing Airplane", June 1970. 81 Tischler, Mark B., "Identification and Verification of Frequency-Domain Models fo r XV-15 Tilt-Rotor Aircraft Dynamics in Cruising Flight", AIAA Journal of Guidance, Control, and Dynamics, July 1986. Tischler, Mark B., "Flying Quality Analysis and Flight Evaluation of a Highly Augmented Combat Rotorcraft", Journal of Guidance, Control, and Dynamics, Vol l4, No. 5, pg 954-964, Sep-Oct 1991. Tischler, Mark B., "Frequency-Domain Identificationof XV -15 Tilt-Rotor Aircraft Dynamics in Hovering Flight", Paper presented at the AIAA/AHSIIES/SETP/DGLR 2nd Flight Testing Conference, Las Vegas, Nevada, November 1983. Tischler, Mark B., "Frequency-Response Method for Rotorcraft System Identification: Flight Applications to B0-105 Coupled Rotor/Fuselage Dynamics", Journalof the American Helicopter Society, Vol 37, No 3, pgs 3-17, July 1992. Tischler, Mark B., "Time and Freiuency-Domain Identification andVerifi cation ofBO 105 Dynamic Models", 15t EuropeanRotorcraft Forum, Amsterdam, September 1989. Wilcox, C.M., "The Effects of Trailing Edge Up Flap Deflectionand Center of Gravity Position on AircraftPower Required and Drag", MS thesis, The University of Tennessee Space Institute, May 1997. 82 APPENDICES 83 APPENDIX A NAVION CALIBRATION CHARTS 84 UTSI Flight Research Navion N66UT Calibrated: 04/05/96 120 i I - l 110 //;z· �/ // 100 --�- + -----�*/ // �<� / -�--:----- ,...... _ �;--- r;n ...... 90 0 /---�/- �/ '-' 12 /- /�/ 70 /- / 60 �� - - -�- ��-- - �- ---- •� ------+--- 50 --�---+-- ---�------··-- 50 60 70 80 90 100 110 120 V 0 (knots) Figure A-1. Copilot's Airspeed Instrument Error Correction N a vi on N 66UT UTSI Flight Research Calibrated: 04/05/96 � 3 �------�------�------�------�------� - � Speed Co urse <> l 2 GPS I )( A X Twr 0 Fly-by - 0 rn � ..... 0 Pace N22UT - -.(>- )C ·� � -- �···· · - 1 � ------__ --- - - Poly. (Speed ----� � -- <> -L - ---- Course) -----r�""K � - --- Poly. (GPS) 0 - -- -r- - - tl . -- � �---..___ <> � �-� ��---- _ � _ l . __ -� 0 • - .Po y (Twr..Fly- by) 00 0\ - u0 <> A t---=< -1 -2 �------+------�------�------r------+------� 60 70 80 90 100 110 120 _ Indicated Airspeed (knots) . - - � Figure A-2. Airspeed Position Error Correction - <> UTSI Flight Research N avion (N66UT) Calibrated: 04/05/96 60 50 40 � ------"-··----- � 30 u <2'-' s:: 20 - _g..... u u 00 -...J 0l:: 10 u 0 -10 -2 0 +----·-t 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 Pressure Altitude (feet) Figure A-3. Copilo t's Altim eter Instrument Error Correction Navion (N66UT) UTSI Flight Research Calibrated: 09/16/98 100 80 60 y=-40.351x + 110.16- 2 R = 0.9998 40 Ol) Cl) Cl I 20 � c; 0 •�-� ' -- 0 -- - J___ -'--- ·-�- 1-- - -- J �- _..L____ _ / ___ > 1 ] t::Cl) 00 -20 00 � 0 u -40 -60 -80 -� -- -100 0 2 3 4 5 6 Reading - Volts Note: +RTAILER ON TE UP -LT Figure A-4. FBW Aileron Stick Position (CHOO) UTSI Flight Research Navion (N66UT) Calibrated: 0911 6/98 15 7.3802x - 19.374 10 y = 0.9997 R2 = 5 01) Q) j_ _ 0 _t___ _ L_ -+-__i_____ L ___!__ _ 1---L . L __ __L_-+ __ .J _ j_ ·f � --" q f t ' Q) ;:j d -5 -- > '"Cl Q) 00 '5 1.0 ;;. � -10 0 u -15 -20 ----� -- -25 �------� 0 0.5 1.5 2 2.5 3 3.5 4 4.5 Reading - Volts Note: +FWD -AFT Figure A-5. FBW Elevator Stick Position (CHOl) Navion (N66UT) UTSI Flight Research Calicrated: 09/1 6/98 3 r------� • y = 1.1575x - 2.7747 2 R2 = 0.9991 1 •··· . -··- -·-· ·· -- -··· --· - .s Q) � > 0 .I--'- _J.__j j- l '-·--' �-- _ L�" L_ + - l_____ J L__ l --+ - l _ L____ L_ '"0 -+ \0 0 i� 0 u -1 1------ -2 .J- -· -3 0 0.5 1.5 2 2.5 3 3.5 4 4.5 5 Reading- Volts Note: + RT PEDAL - LT PEDAL Figure A-6. FBW Rudder Pedal Position (CH02) �------� UfSI Navion (N66UT) Flight Research Calibrated: 09/16/98 4.5 ------, 4 -0.8435x + 4.0178 3.5 y = 0.9999 R2 = 3 .Ei 2.5 u .E"' > 2 ·------'"C u \0 � � 1.5 0 u 0.5 ·· ----- ·· -·-· -··· - -- -··- · 0 -�---�--'---�'- ·-'-·---+ ·- --'- ·-f------____L______j__ ___..1_ __ .___ l ---+--- -- _1______j_ �- - 1 -- ___j -0.5 �------� 0 2 "' 3 4 5 6 Reading - Volts NOTE: +FWD -AFT FigureA- 7. FBWThrottle Command Position (CH03) Navion (N66UT) UTSI Flight Research Calibrated: 09/16/98 20 ------, 15 ----- 10 0 � 5 ·- boCI.I 0 I C1.l 0 l ______L_ _. - -- · ...=('j 1�-�--'--� t- 1- -- > "B 3 2 -- -0.213lx + 1.6542x + 3.4637x - 16.712 -5 �------y = \0 15 N ;; R2 = 0.999 0 u -10 -1------� -15 ---- -�- -20 �------� 0 2 3 4 5 6 Reading - Volts Note: +RT -LT FigureA-8. Roll Rate (p) (CH04) Navion (N66UT) UTSI Flihgt Research Calibrated: 09/16/98 20 r------, 15 • --- -· · - -- - -· 10 ·------ Q) "' 5 ·- - ObQ) 0 ' Q) ::I "Ol 0 -- -+--_L,______j_ ____ l ___ __]______-+ ---.L--�-�· --1-- -______l._.. - > '"!:;I Q) \0 '5 0.1197x3 - 0.9473x2 + 7. 7046x - 14.805 w > -5 y = r:: 0 R2 = 0.9997 u -10 -15 -20 0 2 3 4 5 6 Reading - Volts Note: +NOSE UP -NOSE DOWN Figure A-9. Pitch Rate (q) (CH05) Navion (N66UT) UTSI Flight Research Calibrated: 09/16/98 20 �------. 15 y= -12.013x + 29.521 R2 = 0.9999 10 - --- u 11) "' --- t>J) 5 11) 0 I <1.1 0 _ L_ -�---· --L- __ L ______l______j -1------l______j__ f _____._..__l__ __ -+- ----.....1-�...l - __J ___ .l __ � •.. J_ ._.J __ L__ l __ �> "t:: <1.1 t: 1.0 <1.1 """ > -5 ------s:: 0 u -10 -15 1 �------ -20 �------� 0 0.5 1.5 2 2.5 3 3.5 4 Reading - Volts Note: +NOSE RT -NOSE LT Figure A- 10. Yaw Rate (r) (CH06) Navion (N66UT) UfSI Flight Research Calibrated: 09/16/98 1.50 ..------�------.... 3 2 y = 0.5625x - 0.0359x + 0.5 1 38x - 0.0125 1.00 - ---- 2 R = 0.9966 0.50 ---� fJ) -eo I .2CIS > 0.00 1-. --J ____ L l __ __ L ·-- _l ------�-_..1 __ '"0 � 1.0 -1.00 -1.50 �.-______...... ______-1.5 -1 -0.5 0 0.5 1 1.5 Reading - Volts Note: +FWD -AFT FigureA- 11. Longitudinal Acceleration (ax) (CH07) Navion (N66UT) UTSI Flight Reasearch Calibrated: 09/1 6/98 1.50 ------""!"'------. 1.00 • �� -- -� -� -- �� �------·- � � 0.50 -!:>0 ' -1.00 -1.50 ..______....,______. -3 -2 -I 0 2 3 Reading - Volts · Note: +RT -LT Figure A- 12. Lateral Acceleration (ay) (CH08) Navion (N66UT) UTSI Flight Research Calibrated: 0911 6/98 1.50 .------�------. 1.00 �-· -- -� y = -0.61 16x3 - 0.0229x2 - 0.5084x - 0.0058 R2 = 0.9959 0.50 "' -00 I > 0.00 ______j______l______..___ _j_ ------'------4 -- _j______J._ ----· __ .!,.___ • _L "0� \0 -..J � u § -0.50 1------·- �------ -1.00 1------ -1.50 -1.5 -I -0.5 0 0.5 1.5 Reading - Volts Note: +UP - DN Figure A-13. Normal Acceleration (az) (CH09) Navion (N66UT) UTSI Flight Research Calibrated: 09/1 6/98 25 • 20 -- 2 y = 0.7397x + 4.1392x - 16.968 2 15 R = 0.9979 10 bl) v Q I v ::s 5 � > "'Cl v 0 ___.J------+--�-1--·- _l _. _ L . +------L-- ___ I ____i_ _ _l_ \0 t:: 00 v � 0 u -5 -10 -- - -15 ---�----� -----· -20 0 1 2 3 4 5 6 Reading - Volts Note: + Right Aileron TEUP - Right Aileron TEDN Figure A- 14. Aileron Surface Position (CHlO) UTSI Flight Research Navion (N66UT) Calibrated: 09/16/98 40 �------� 30 • y= -10.337x + 33.062 2 R = 0.997 20 01) 0 Q I 10 0 E«< > '"0 0 \0 t: 0 • -�-�' "--- �-' � I � � - - , - � \0 0 + I + :> d 0 u -10 -� -20 I ------�- ·- -- �------ -30 �------� 0 2 3 4 5 6 Reading - Volts Note: +TRAILING EDGE UP - TRAILINGEDGE DN Figure A-15. Elevator Surface Position (CHll) UTSI Flight Research Navion (N66UT) Calibrated: 09/16/98 30 25 20 15 y = 8.0966x - 17.234 bl) 10 - --�- 1------C1) 1 2 0 R = 0.9983 I C1) 5 ..E"' > - "'CC _ �L- .-- ___ L______l ______L______L l___ _ 0 C1) 0 .. - t------'-----+ -� 0 t:: C1) � 0 -5 u -10 -15 -20 -25 0 1 2 3 4 5 6 Reading - Volts Note: +RT -LT Figure A-16. Rudder Surface Position (CH12) Navion (N66UT) UfSIFlig ht Research Calibrated: 09/16/98 4.5 4 3.5 ··------ y = -0.868lx + 3.977 R2 0.9975 3 = 1----- .Ei I 1 ------·-- 0.5 0 0 0.5 1.5 2 2.5 3 3.5 4 4.5 5 Reading - Volts Note: +FWD -AFf Figure A-17. Std. Throttle Control Position (CH13) UTSI Flight Research Navion (N66UT) Calibrated: 09/1 6/98 60 .------� 40 20 y = -32.869x + 81.584 R2 =I t>O Q) 0 I 0 +- -i Q) I "--+--"--'-----'--�- � > "0 - Q) t:: -20 --�- ---· 0 Q) 1- N ;> 1::: 0 u -40 - ··------��------·------t -60 - I - -80 �------� 0 0 5 1.5 2 2 5 3 3.5 4 4 5 5 . . . Reading - Volts Note: +RT -LT Figure A-18. RollAttitude (�) (CH14) Navion (N66UT) UTSI Flight Research Calibrated: 09/16/98 ''• 50 40 30 20 00 u y = l6.476x - 41.064 0 I 10 I -- - R2 = l u ::s c;j > 0 - "0 0 u w t: u > -10 d 0 u -20 -30 -40 -50 0 2 3 4 5 6 Reading - Volts Note: +NOSE UP -NOSE DN Figure A-19. Pitch Attitude (8) (CH15) APPENDIX B CIFER BODE PLOTS 104 LJ.,.t..t< v. SfLONSWP_COM_�HCU�_LONS_Q g: c mot "' 1: ;��::: __�:-�=s 18I' L - ..,.--- 101 t --. - ..-----.- 1.0" a --.- u .,..--,--,--�� .nc �Freq uenc� CRad/sec l .. :: .. 'i' "'-... .I.: I n. - 0 ------� VI � !I' I 1 10-1 11!1 " 11!1 ( Rad/s"'c l - --- .,q_, _':::nc\1 ....------F_r__ � 01 .... u c ""' L . ., .. .£: Q u .. .. "' ··r·---r-- --r-r--r- "'' ·T-···- '+------..-----.----.--.,--.,--,r--r--.--,------.,----t -.· ll'l- HI" 10' Fr.,quen cy CRad/sec l Figure B-1. Conventional Control Longitudinal Input To Pitch Attitude Response ------····------ ... �\ : .,; -r- z: 0 i ..... :1 \ U...... N ,-....._ w.,; .J lo... "" w 0 � 1JI ; .J N I• \ I 0. "'"' ...... z: 0 !I. u I, I � J ... I of I I I I I .. z I 6 ? 8 .. l 3 I .. 5 9 TIME< SEC > I - .. 0 "' VI w /� tn z: 0 a.. i tn I \ LoJ I "' 'I/ I I - \ --...... I � ····· v ·· ······· v ········· / ··· ···· I ··•········ v � ! I ! I "'I I .. ��-,- a l z 3 .. 5 6 7 8 9 TIME Figure B-2. Conventional Control Longitudinal Doublet FWLONSWP_COM_ABCD0_LONS_Q .. N c """ " E :=-=-�===--=---�� N.. ' r-· -----r-----·---r,---,��...,--, - 1 0-1 10" H l l (Rad/sec l .. Frequency ------!' .. " ..c ------�------ll.. i 'l' ------=---=.: - 0 -.) I illI 1 111" 1 HI" llil' Frequenc!:l IRad/sec l --v CD .,.,;�l 0 c .,.., '-. GJ"' .r: 0 u.,. .. ... ,.; l 1111-1 10" t f.l Frequency IRad/sec ) Figure B-3. FBW Control Longitudinal Input To Pitch Attitude Response Ill c=== re =:::::c cc m ==> ��- :-�c- w --� Q ::::J O> f- - . z�...... -- ��--- '-'' - -- a: --- E:� ���-:::-:_--�� � "' '? ------�------� --=------. . L'> w-� - o.., ------..______-on W;' (Jl � a: --- 6:,"' I � ...... 0 00 - I ':?"'" -���------� w u z: w.., �· w "' :r 0 u "'"! 11'!- 1 tB" Hl1 lB" FREQUENCY STLONSWP_COM_ABCD0_LONS_Q Pw : -1 FWLONSWP_COM_A BCD0_LONS_Q Ph : 360 .00 Pw: -1 Figure B-4. Conventional and FBW Longitudinal Control Input to Pitch Attitude Response � ------=: -:-:-=��:�::.:-.:-·:=�·-:.::-::::::::-:-::�::- m ill "' w 0 ::J"' r"' H z t!l O:"' :E � .. � (aW I "' w �------� til Q._l�� ll! r ...... 0 \0 - I - ...., __,.,, ---, '� w _ _.,. u ,. �'-..._ ), z: w ... 0<• w "' I 0 u N .. H!l-1 1 1110 0 1 ' 2 FREQUENCY (RAD/SEC> 1 0 STLONSWP_COM_ABCD0_LONS_AZ FWLONSWP_COM_ABCD0_LONS _AZ Figure B-5. Conventional and FBW Longitudinal Control Input to Heave �.. mre (:.) · ,,, ------·- ·-:--..:.::··- --- n =>• ....------�� -::-:--:-.::=:-:::::-��:�., . ' I' 1- � .... ' 7. v ... ' • J ' '' ,,�• .. \ .,).� �I:• Ill "-----'<..(�/ ' ' ' ,, � .. "' 'I I .. I I I I I � 'I'' I p I - .. ' I.J)on I ·� n ' r. • l I w 11\t I 11 :1 I I" llI ��-· ' "'I I w =-= � ll'r-/1I I I 1.11 ' =----=Y<-c-=.c- f\J I IJ I cr• _ ! r :r"' _, , ,,l1 ,,I It'' !''I o.. O' '"" r. v 1 ' .. "'\..,...r � ' N"' .. .. '1' l - - I 0 - - - , I / \ " w I \ u z I \ w., � I \I 0:: · I w'" '-./ :r \ I II 0 \./ 1\ LJ \..' "' .. 1 1 1 0- 1 0" 10 1 02 FREQUENCY STLONSWP_COM_RBCD0_LONS _P Ph : 360 .00 FWLONSWP _COM_RBCD0_LONS_P Figure B-6. Conventional and FBW Longitudinal Control Input to Roll Rate Response SIL�ISW�_COM_RBCD�_LA IS_P .. ' -�------ c: >::� � � � 1 0-1 1 08 10' 1�-�- Fre-�q uency CRad/sec ) �� ------··------'".. z .<:•� n. ' - - - I � -t 108 10' 1� Fr �quenc� CRad/sec > � � -�� "! .. ., u c: ., ., �m J:: 0 u . .. .. .,; Figure B-7. Conventional ControlLateral Input To Roll Attitude Response - :1: ! ... -- r-. N �---- \ --- ..,--- I I \ ------·-··· -,- - I i!!S !D. I I I u.lt; I \ u.l� ... c::::) � - a::cS ' !!! I I- I I I � -- --·- - I � (.,.)lil I I -- : ... I!!• 1 2 " I " 6 7 a " --; .. I TIME .,. - - I N � - 1--- � ... /-- � ... L&.l ' � l( "' :z:: �'f ··""' / t3 ' 0::: 7! '\ �/ � ' ... l!l.. 1 2 3 " 6 7 a " 18 TIME< SEC > Figure B-8. Conventional Control Lateral Input to Roll Attitude Response Doublet FWLRTSWP_COM_RBCD0_LA1S_P �I ri -� m � +-----�--���----�--��------�--�� 1 0 -1 10" 1 01 Frequenc� CRad/sec l !-I .. "' .s::. a.. N"' I - w I • 10-l 1 0" 1 0' Frequency (Rad/sec l ID-1 uIll .. c IJu:o <.. IIJCD .s::. 0 u .. .. 1\J .,; 1 0-1 1 01!1 1 01 Fr equenc� CRad/sec l Figure B-9. FBW Control Lateral Input To Roll Attitude Response � 111 "' 0 w •··3:S I �� ----� �-- z:.... - �'>.:• ------�"------ ' � :c == _ -�I -7 _ _" _ ' ..,· ---"- I.J ---::-�------:C::::::� e :�� .... � tn' ([ ifa - 1 'l' ._ .1' , � - - I � I w u z: w., �- w • I 0 u "' .. 1 0 · 1 10" 101 102 FREQUENCY STLATSWP_COM_ABCD0_LATS_P Pw : -1 FWLATSWP_COM_ABCD0_LATS_ P Pw : -1 Figure B-1 0. Conventional and FBW Lateral Control Input to Roll Attitude Response :::::: � �=-cc---:o ,_..=o-::- .. .-o._-:c;·�c-.=-�.,.__"-' �� LoJ "'"' '·"'"'-... :::J • �""- 1-' H z CJ,. � ··..,.---,- .. � "'.. ...� N ==--=--.::..-=-��------��- -.;:, " Cl' w �!Ul ' I w w!£ .. "'" LoJ "' I 0 u N oi 1 1'!-l t ee 1 1'! ' 1 1'!2 FREQUENCY CRAD/SECl STLATSWP_COM_RBCD0_LATS R Pw : -1 FWLATSWP_COM_ABCD0_LRTS_R Pw : -1 Figure B-11. Conventional and FBW Lateral Control Input to Yaw Attitude Response ;;; C'l $ "'wt;J ::J ·-- .. � � Z"' =:-::::-=--�·-=--...::.:-·:::::-..:�:::.�� " �T "' -�' E .. �_7+-,.___c�'l.....�:;'�'J. 'f � � o�a:"' -· �-=====-�, 1-J{za:� ',� :r:' !L II ,-:_.'-" " � -- 1 \ � tf:o - 'f ''·�' - I I 0\ -----r-��'-- I 1 \ I '-1 '�' I hi u z: w .. C>:· ,._, .. :X: 0 u N.,; H I-t 10 10' 02 '" FREQUENCY (RAD,SEC J 1 STLATSWP_COM_ABCD0_LATS_AY FWLRTSWP_COM _nBCD0_LnTS_nY Figure B-12. Conventional and FBW Lateral Control Input To Lateral Acceleration SfPEDSWP_COM_RBCD8_PED_R g: � �-�------�- c C') .... r7 �11"! _,���::==;=:�� Hl 0 11<1 1 Fr Equency (Rad/sec ) r .. z� .r:• IL ' ...... :...... --J ;··=... · ·.···,··:· ...,...... ,.. .. � , ...,.�,., ,,=:;=� I I I I I I I u;�t 1 a- 1 lB" �I Frequency '" .,..; u c .. ... L • ...... r: 0 u"; "' N "'1------.------.----.----.--.--.--.-.-,------.------.-----.---.---r---.--.-.-.------.------.----.----.---.-.--�,- ° 1 1 0-1 1 0 1 0 Fr equency (Rad/sec l Figure B-13. Conventional Control Pedal Input to Yaw Attitude Response FWPE.DSWP _COM_RHLD0_J'LD_f-< 11! --.....-- -- � � -· � c m ., .. I:i ' �-::�·- 10- • 1 1'19 1 0 1 Frequency. �lID .. .. r u c ...... c . 0 u .. .,; N .,; ll t e- 1 Uil 10' Frequen cy CRad/sec ) Figure B-14. FBW Control Pedal Input to Yaw Attitude Response ill ------.,-'""--.::..-, _�:--._ m�0 --.. --= -�"� -- ...... w 0 :r> -/ ::l"' "'<"� "-- I- ....z "' � -- �·o- ; �· "'- ��a-- '- ::t:' �<- - - 0.. �-:::-:::.-=--� �- - � - - - - - r �J'-'t � � !XII - - \0 I t.J !i 0!·,., .. w "' ::t: 0 u .. .; ' 1 1!-1 10. tl'l 10" FREQUENCY (RAO/SEC l STPEDSWP _COM_ABCD0_PED_R Pw : -1 FWPEDSWP_COM_ABCD0_PED_R Pw : -1 FigureB- 15. Conventional and FBW Pedal Control Input to Yaw Attitude Response !1: - .. --·�---- '". en� n w 0 :::J a>· ' 1- , H z e>., --�-=�<-::_��'�, '/.--: ��� ��----�--���----�--���--��--��--� Ii"' · I • ==------I I wG-: I I �� I bl• I !j I IL. I - +------�------�--��--�-.--.--T-.-.r------�------�-----T---,---T--�-r-T�------�------T-----r--�---r--�o--r-, N ' 0 � � w"'�"! :I: 0 u N • +------,------�----r---�--r--r-,--r-r------,------�r---�----��--,--r-,-,r------���------,-----r---,---T--�-r-r-, 10-1 llll" 10' 1 00! FREQUENCY STPEDSWP_ COM_RBCD0_PE D_P Pw : -- 1 FWPEDSWP_COM_RBCD0_PED_P Pw : -1 Figure B-16. Conventional and FBW Pedal Control Input to Roll Attitude Response � ;a� r:. w CJ -� -:-=:_:::" ::::1 .. =-...z::.-=...... :::::-:-=----=---=-� 1- H .- T -�-.-<- '""'-'"--o- ���� �!!1.:' 1!1 Ill • :::::-.::::-::--:::::-::::-=::- <;:r: �·-· w ·• Ill ' ill !f ...... n.i �- - ��...... � --- ._, J.J "'-- ' --. ,..,,-.- _,-,,,.(-\}, ' - I N - \.1 �� ------=--=�_,�-��-...... "\. I w ,.\ u , I z: w.., I I "' . I \I I �0 .. I u I �I l • "'' 't N ' v \ 1 oi ,--·-,---.-,------.----. ..---..--. .---.--,----.--.--r-----.:..L...----..---....----.--,---,--,-. , 1 lli!- te• te• 1 02 FREQUENCY CR�D,.SEC J STPEDSWP_COM_ABCD0_PED_AY FWPEDSWP_COM_ABC00_PED AY Figure B-17. Conventional and FBW Pedal Control Input to Lateral Acceleration w iD .-::::: / \ �re / - pc w -- ,/ � -- ' y-v.. 0 / -- � '-1 :::1 �-� ,, t- � 't �-a: >: • 'l' 1111 1 ! II ' 1 ,I ill I I · �ii ------� I '-" -=-=..r"' - 1 w � r 0 / 1 �. -- I �I v I w '- 1 Ul I a: \ ) I I \ /' I ' ' I \ I � 1 , , \ I fi:� '/ lo�� I - I � N N I w u z ...... 00· ..� . :r: 0 u .. •. te-• t0• 101 102 FREQUENCY STTHRSWP_FRE_000D0_THR_RZ Ph : 360 .00 FWTHRSWP_FRE_000D0_THR_�z Ph : 360.00 Figure B-18. Throttle Control Input to Longitudinal Acceleration VITA Randy Bolding was born in Mason City, Iowa on August 23, 1957. He attended schools in the public systems of Oklahoma City, Oklahoma, and Manly, Iowa, where he graduated from North Central High School in May 1975. He entered the United States Army in February of 1977 as a Military Policeman. In September 1981, he completed Rotary Wing Flight Training at Ft. Rucker, Alabama and was commissioned as a Warrant Officer. He completed an Associate of Science Degree in Pre-Engineering at Methodist College, Fayetteville, North Carolina, in 1986. He is a veteran of the Operation Urgent Fury/Grenada and Operation Desert Shield/Storm. He served a total of fourteen months working drug interdiction in the Bahamas fo r the State Department. Through use of continuing education programs, he was able to complete a Bachelor of Science degree in Business from the University of Maryland in 1994 and a Master of Science Degree in Engineering Technology from Murray State University, Murray, Kentucky in 1995. In August of 1997, he attended The University of Tennessee Space Institute, graduating in December 1998 with a Master of Science degree in Aviation Systems with a follow on assignment to attend the U.S. Naval Engineering Test Pilot School at Patuxent River, Maryland. 123